Directional reference, discourse, and landscape in Ahtna

UNIVERSITY OF CALIFORNIA

Santa Barbara

Directional Reference, Discourse, and Landscape in Ahtna

A Dissertation submitted in partial satisfaction of the

requirements for the degree Doctor of Philosophy

in Linguistics

by

Andrea Lauren Berez

Committee in charge:

Professor Marianne Mithun, Chair

Professor Sandra A. Thompson

Professor Patricia Clancy

Professor James Kari

June 2011

The dissertation of Andrea Lauren Berez is approved.

____________________________________________ James Kari

____________________________________________ Patricia Clancy

____________________________________________ Sandra A. Thompson

____________________________________________ Marianne Mithun, Committee Chair

June 2011

iii


Copyright © 2011

by

Andrea Lauren Berez

iv

ACKNOWLEDGMENTS

There are many people who deserve recognition and gratitude for their help with

completing my degree. First among them is Mr. Markle Pete, who has been my

primary Ahtna teacher since 2008, and who shared his language, his time, his values,

his food, his stories, and his family with me. Tsin’aen, Markle.

Other Ahtna elders also shared their language and their knowledge with me, and

I thank them here: Mrs. Virginia Pete, Mrs. Jeannie Maxim, the late Mr. Luke

Maxim, Mrs. Ina Lincoln, Second Chief Fred Ewan, Chief Ben Neely, Mrs. Hazel

Neely, the late Chief Jim McKinley, the late Mr. Adam Sanford, the late Mr. Jake

Tansy, Mrs. Katie John, and the late Mrs. Martha Jackson.

Many people helped me feel at home in the Copper Valley. My professional

welcome was facilitated in two ways. First, in 2007 the teachers at the Ya Ne Dah

Ah School, Kari Shaginoff, Daniel Harrison, Jesse Boger, and the late Mrs. Katie

Wade, invited me to visit them in Chickaloon from Kenai and first suggested I spend

time with their respected Elders; I am grateful for this opportunity. Second, the

visionaries at the Ahtna Heritage Foundation and C’ek’aedi Hwnax gave me an

unexpected purpose and quickly became my treasured friends and colleagues: Taña

Finnesand, Karen Linnell, and Liana Charley John.

My personal welcome to ’Atna’ Nene’ was made warm by many other people:

Gloria Stickwan, the late Cathy Dewitt, Mrs. Virginia Pete, Tana Mae Pete, Kayla

Pete, Elizabeth Vollema (the formidable Stickwan women); Timmy Pete and

Brandon Pete; Taña and Martin Finnesand; Michelle Bayless; Todd Roeske and the

v

folks at Mt. Drum Lutheran Church; Glenda Sinyon; Ann Urich; Travis Goodlataw,

Lishaw Lincoln, Susan Voyles, Mark Johns, and Starr Knighten.

Academic gratitude goes to the members of my dissertation committee. Marianne

Mithun has offered unflagging support in all matters professional and linguistic since

I arrived at UCSB in 2006, and I am grateful for her mentoring. Sandy Thompson

provided valuable feedback on this and other work and has been generous with her

friendship as well. Pat Clancy’s feedback here was invaluable for pointing out the

implications of the interactive nature of my work with Ahtna consultants. Jim Kari

deserves my thanks for his openness with his data and for paving the way for me in

both the Dena’ina and Ahtna communities by introducing me to his many old

friends. Students of Alaskan Athabascan languages owe him a debt of gratitude for

his years of dedicated scholarship.

Professional support has come from others as well: Helen Aristar-Dry, Anthony

Aristar, and Gary Holton all facilitated my first work in Alaska via the Dena’ina

Archiving, Training, and Access project; Carol Genetti has encouraged my

development as a documentary linguist; Jean Mulder has been a friend and has

taught me many things about academic collaboration. Many thanks to you all.

Finally there are personal thank-yous to extend. To Hope, who helped me get

organized. To Dov, who came along at the end and kept me focused on the goal. To

my Short List (in chronological order): Jennifer Hickam, Gary Harmon, Lou

Marinaro, Martha Ratliff, and David Miliotis. To Ayla Bozkurt Applebaum and

Uldis Balodis for their companionship over the last five years. And finally to my

vi

family: my parents Ellen and Bill and my sister Leslie, who buttressed me through so

many years of school and in the face of many changes.

______________________

Funding for the research presented here was provided by the University of

California Pacific Rim Research Program, the American Philosophical Society, the

Jacobs Fund, the University of California Humanities Research Institute, the UCSB

Graduate Humanities Research Program, the UCSB Department of Linguistics, and

an NSF grant to Marianne Mithun (NSF ASSP #0853598).

vii

VITA OF ANDREA LAUREN BEREZ

Appointment

August 2011- Assistant Professor, Department of Linguistics, University of

Hawai‘i at M noa.

Education

2011 PhD Linguistics, University of California, Santa Barbara.

2006 MA Linguistics, Wayne State University.

1998 MFA, Sculpture, New York Academy of Art.

1994 BFA, Sculpture, University of Michigan.

Fellowships and Awards

2011 Isaak Walton Killam Postdoctoral Fellowship, University of Alberta

(declined).

2010-2011 Graduate Fellowship in the Humanities, University of California.

2009-2010 Graduate Humanities Research Program Fellowship, UC Santa

Barbara.

2007-2008 Graduate Opportunity Fellowship, UC Santa Barbara.

2006-2007 Graduate Opportunity Fellowship, UC Santa Barbara.

2006-2011 Linguistics Department Fellowship, UC Santa Barbara.

2006 Connaught Scholarship, University of Toronto (declined).

Publications

Edited volumes: 2010 Berez, Andrea L., Jean Mulder, and Daisy Rosenblum (eds.).

Fieldwork and Linguistic Analysis in Indigenous Languages of the Americas. Journal of Language Documentation & Conservation Special Publication No. 2. Honolulu: University of Hawai‘i Press.

2007 Kari, James and Andrea L. Berez (eds.). Inland Dena’ina Keywords:

Lime Village Dialect, by Helen Dick. Anchorage: Alaska Native

Heritage Center.

2007 Berez, Andrea L., Suzanne Gessner, Leslie Saxon and Siri Tuttle

(eds.). Alaska Native Language Center Working Papers 6: Working

Papers in Athabascan Linguistics. Fairbanks: Alaska Native Language

Center.

Refereed publications

Journal articles:

To appear Berez, Andrea L. Intonation as a genre distinguishing feature in Ahtna:

A quantitative approach. Functions of Language.

viii

2010 Berez, Andrea L. and Stefan Th. Gries. Correlates to middle marking

in Dena’ina iterative verbs. International Journal of American

Linguistics 76(1): 145-165.

Refereed book chapters:

To appear Thieberger, Nicholas and Andrea L. Berez. Linguistic data

management. In Nicholas Thieberger (ed.), The Oxford Handbook of

Linguistic Fieldwork. Oxford: Oxford University Press

2006 Berez, Andrea L. and Gary Holton. Finding the locus of best practice:

Technology training in an Alaskan language community. In Linda

Barwick and Nicholas Thieberger (eds.), Sustainable Data From

Digital Sources: From Creation to Archive and Back, 69-86. Sydney:

Sydney University Press.

2006 Holton, Gary, Andrea L. Berez and Sadie Williams. Building the

Dena’ina language archive. In Laurel Evelyn Dyson, Max Hendricks

and Stephen Grant (eds.), Information Technology and Indigenous

People, 205-209. Hershey, PA: Idea Group.

Refereed conference proceedings:

2009 Applebaum, Ayla B. and Andrea L. Berez. A theory is only as good

as the data: Casting a wide net in Kabardian and Ahtna

documentation. In Peter K. Austin, Oliver Bond, Monik Charette,

David Nathan and Peters Sells (eds.), Proceedings of Conference on

Language Documentation and Linguistic Theory 2, 29-38. London:

School of Oriental and African Studies.

2009 Berez, Andrea L. and Stefan Th. Gries. In defense of corpus

linguistics: A behavioral profile analysis of polysemous get in English.

In Steven Moran, Darren S. Tanner and Michael Scanlon (eds.),

Proceedings of the 2008 Northwest Linguistics Conference. Seattle:

UW Department of Linguistics. Online:

http://depts.washington.edu/uwwpl/editions/vol27.html.

2007 Berez, Andrea L. Middle marking in Dena’ina iterative verbs. In

Berez, Andrea L., Suzanne Gessner, Leslie Saxon and Siri Tuttle


Papers in Athabascan Linguistics, 21-39. Fairbanks: Alaska Native

Language Center.

2006 Berez, Andrea L. and Gary Holton. Finding the locus of best practice:

Technology training in an Alaskan language community. In Linda

ix

Barwick and Nicholas Thieberger (eds.), Sustainable Data From

Digital Sources: From Creation to Archive and Back, 69-86. Sydney:

Sydney University Press.

2006 Holton, Gary, Andrea L. Berez and Sadie Williams. Building the

Dena’ina language archive. In Laurel Evelyn Dyson, Max Hendricks

and Stephen Grant (eds.), Information Technology and Indigenous

People, 205-209. Hershey, PA: Idea Group.

Reviews:

2010 Berez, Andrea L. Review of Tanacross Learners’ Dictionary:

Dihthâad Xt’een Iin And eg’ Dínahtl a’, compiled by Irene Arnold,

Rick Thoman, and Gary Holton, edited by Gary Holton. Fairbanks:

Alaska Native Language Center. International Journal of American

Linguistics 76(3): 397-400.

2007 Berez, Andrea L. Software review: EUDICO Linguistic Annotator

(ELAN). Journal of Language Documentation & Conservation 1(2):

283-9.

2003 Berez, Andrea L. Book notice for Multilingualism in Italy Past and

Present, by A. Lepschy and A. Tosi. Diachronica 20(2): 386.

Non-refereed and other publications:

2010 Daisy Rosenblum and Andrea L. Berez. Introduction: The Boasian

tradition and contemporary practice in linguistic fieldwork in the

Americas. In Andrea L. Berez, Jean Mulder, and Daisy Rosenblum

(eds.). Fieldwork and Linguistic Analysis in Indigenous Languages of

the Americas. Journal of Language Documentation & Conservation

Special Publication No. 2, 1-8. Honolulu: University of Hawai‘i

Press.

2009 Holton, Gary, Andrea L. Berez and Sadie Williams. Dena’ina

Archiving, Training and Access. UNESCO Register of Good Practices

in Language Preservation. Paris: UNESCO Endangered Languages

Programme. Online: http://www.unesco.org/culture/ich/index.php.

2007 Berez, Andrea L. Middle marking in Dena’ina iterative verbs. In

Berez, Andrea L., Suzanne Gessner, Leslie Saxon and Siri Tuttle


Papers in Athabascan Linguistics, 21-39. Fairbanks: Alaska Native

Language Center.

x

2005 Dena’ina Language Archive and Dena’ina Qenaga website. Online:

http://qenaga.org. [with Gary Holton and Sadie Williams]

2005 E-MELD School of Best Practice website: From cassette to easy-

access software: Dena’ina. Online:

http://emeld.org/school/case/denaina/index.html.

2004 E-MELD School of Best Practice website: Shoebox legacy data:

Mocoví. Online: http://emeld.org/school/case/mocovi/index.html.

xi

ABSTRACT


by

Andrea Lauren Berez

This dissertation examines one corner of the grammar of the Ahtna Athabaskan

language of Alaska: the use and semantics of the lexical class of directionals. In

particular, this dissertation looks at how Ahtna speakers use directionals in

spontaneous discourse and elicitation against the backdrop of the physiography of

Ahtna territory. The semantics of the directional system is traditionally riverine,

meaning that the orientation of the local river local determines which directional

term speakers choose. Talk about direction and location of referents in the natural

landscape is common among Ahtna speakers: Ahtna people are traditionally semi-

nomadic, and verbally displaying one’s knowledge of overland travel through Ahtna

territory has a special place in culture and society. Among the linguistic resources

available for describing concepts like path and location are the directionals, the use

of which is a direct reflection of a speaker’s familiarity with the geography of the

region he or she is describing. Awareness of the local ecology is thus not only

central to Ahtna cultural practices, but also potentially influences the development of

the grammar over time.

xii

This dissertation is concerned with the relationship between language change

over time and the use of the directionals in discourse and elicitation. The first section

examines the recent changes in the semantics and usage of the directionals because

of language contact. Using data from my fieldwork, I show that the nearly constant

contact of Ahtna with the dominant English language is causing a shift in the

semantics of the directionals, such that they now refer less to the orientation of the

local river, and more to the cardinal directions found in English.

The second section looks at language change from a purely language-internal

point of view. Using data from a previous generation of Ahtna speakers, it discusses

how the complex morphology of directionals is lexicalizing over time, leading to a

loss of semantic clarity that speakers are compensating for via other resources in the

discourse structure. Central to the discussion is the topography of the landscape

itself. To that end, geographic information systems technology plays a large role in

the data presented here.

xiii

TABLE OF CONTENTS

Acknowledgments .................................................................................................. iv

Vita........................................................................................................................ vii

Abstract .................................................................................................................. xi

Table of Contents..................................................................................................xiii

List of Examples................................................................................................... xix

List of Figures..................................................................................................... xxiii

List of Tables....................................................................................................... xxv

List of Maps....................................................................................................... xxvii

Chapter 1: Introduction ............................................................................................ 1

1.1 Objective of This Dissertation........................................................... 1

1.2 The Linguistic Setting....................................................................... 3

1.3 Genetic Affiliation............................................................................ 5

1.4 Typological Sketch ........................................................................... 9

1.5 Review of Previous Research on Language and Geography in Ahtna,

Athabascan, and General Linguistics .....................................................14

1.5.1 Kari’s Work on Geography and Place Names in Ahtna and

Athabascan .....................................................................................14

1.5.1.1 Lists of Toponyms.........................................................15

1.5.1.2 Collections of Travel Narratives ....................................15

1.5.1.3 Analytical Discussions for Linguistic Audiences ...........15

xiv

1.5.1.4 Reports for Nonlinguistic Audiences .............................16

1.5.2 Other Sources on Na-Dene Geolinguistic Knowledge.............16

1.5.3 Ahtna Ethnography ................................................................17

1.5.4 Descriptions of Directional Systems in Northern Athabascan .17

1.5.5 Spatial Cognition in Language ...............................................17

1.5.6 Route Description in Discourse ..............................................18

1.5.7 Spatial Concepts in Narrative .................................................19

1.6 Organization of This Dissertation....................................................20

Chapter 2: Background on Directionals..................................................................23

2.1 Spatial Frames of Reference in Ahtna Grammar and Lexicon .........23

2.1.1 Ahtna Intrinsic Frame of Reference........................................26

2.1.2 Ahtna Relative Frame of Reference........................................28

2.1.3 Ahtna Absolute Frame of Reference.......................................29

2.2 Morphology of the Directionals ......................................................30

2.3 Semantics of the Directional Stems.................................................32

2.4 “Major River Orientation, Principles 1 and 2” of the Axis of the

Riverine Spatial Grid............................................................................34

2.5 Conclusion......................................................................................41

Chapter 3: Contact-induced Semantic Change in Ahtna Directionals......................43

3.1 The Ahtna Region...........................................................................44

3.2 The Bilingual Fieldwork Conditions ...............................................46

xv

3.3 Forming the Hypothesis ..................................................................51

3.4 Evidence for Change: nae’ ‘upriver’ Becoming Equated with

‘north’ ..................................................................................................56

3.5 The Role of General Topography....................................................67

3.5.1 Ngge’ ‘upland’ in the Matanuska River Drainage ...................67

3.5.2 Tgge’ ‘up (vertically)’ Along the Tazlina River to Tazlina

Lake ...............................................................................................69

3.5.3 Why ngge’ and tgge’? ............................................................71

3.6 Conclusion......................................................................................78

Chapter 4: Refining the Semantics of nse’ ‘ahead’ .................................................80

4.1 Semantic Complexity of nse’ ..........................................................80

4.2 Possible Meanings ..........................................................................83

4.2.1 Meanings Arising From Fieldwork.........................................83

4.2.1.1 Further Away Than ’un’e ..............................................83

4.2.1.2 Coastal Locations ..........................................................87

4.2.1.3 Over “The Hump of the World”.....................................96

4.2.2 Other Possibilities From the Literature .................................103

4.2.2.1 Association With the Arc of the Sun as a Metaphor for the

Hearth.....................................................................................103

4.2.2.2 A Lexicalized Toponym ..............................................105

4.3 Toward a Unified Definition of nse’..............................................106

4.4 Conclusion....................................................................................110

xvi

Chapter 5: Directional Reference in Discourse and Narrative: Comparing Indigenous

and Non-indigenous Genres .................................................................................113

5.1 Frog Stories and Travel Narratives in the Study of Direction and

Location .............................................................................................114

5.2 Grammar of Direction and Location in Saen Tah Xay Tah, Ta’stede

Dze’, and Naghaay .............................................................................121

5.2.1 Derivational/Thematic Adverbial Prefixes............................122

5.2.2 Directionals in Discourse .....................................................126

5.2.3 Toponymy............................................................................129

5.3 Attention to Narrative Tasks .........................................................132

5.3.1 Discourse Use of the Postpositional Phrase yihwts’en...........134

5.3.2 Discourse Use of the Adverb xona .......................................137

5.3.3 Tracking Referents with Relative Clauses ............................140

5.4 Conclusion....................................................................................143

Chapter 6: Lexicalization of the Directional Prefixes: Evidence from Discourse ..146

6.1 Introduction ..................................................................................146

6.2 Directional Morphology in Discourse ...........................................149

6.2.1 Data and Methodology .........................................................152

6.3 Historical Sources of the Directional Morphology.........................156

6.3.1 Historical Sources of the Directional Prefixes.......................157

6.3.2 Historical Sources of the Directional Stems..........................158

6.3.3 Historical Sources of the Directional Suffixes ......................159

xvii

6.4 Compositionality in the Directional Morphology ..........................161

6.5 Mapping the Use of Deictic Prefixes in the Travel Narratives .......167

6.5.1 The Distal Prefix ’u-.............................................................167

6.5.1.1 Mr. Sanford’s Use of the Distal Prefix ’u-....................167

6.5.1.2 Mrs. John’s Use of the Distal Prefix ’u- .......................173

6.5.1.3 Mr. Tansy’s Use of the Distal Prefix ’u- ......................179

6.5.2 The Medial Prefix na- ..........................................................184

6.5.2.1 Mr. Sanford’s Use of the Medial Prefix na- .................186

6.5.3 The Proximal Prefix da- .......................................................192

6.5.3.1 Mr. Sanford’s Use of the Proximal Prefix da- ..............192

6.5.3.2 Mrs. John’s Use of the Proximal Prefix da- .................194

6.6 Discussion: Lexicalization in the Prefixes .....................................198

Chapter 7: Directional Suffixes and Discourse Structure ......................................202

7.1 Introduction ..................................................................................202

7.2 Literal Use of the Directional Suffixes ..........................................203

7.2.1 The Punctual Suffix -t ..........................................................203

7.2.2 The Area Suffix -xu..............................................................204

7.2.3 The Allative Suffix -e...........................................................205

7.2.4 The Ablative Suffix -dze ......................................................206

7.3 Discourse Use of the Directional Suffixes.....................................207

7.3.1 Discourse Use of the Punctual Suffix -t ................................207

7.3.2 Discourse Use of the Area Suffix -xu ...................................210

xviii

7.3.3 Discourse Patterning of Both Suffixes..................................214

7.4 Reclaiming Distance Distinctions Via Discourse Use of

Suffixes ..............................................................................................222

7.5 Conclusion: Grammaticalization in the Suffixes............................223

Chapter 8: Summary ............................................................................................226

Appendix: Transcription and Glossing Conventions.............................................230

References ...........................................................................................................236

xix

LIST OF EXAMPLES

Example 1.1. Noun and verb stems from same root, Slave ..................................... 13

Example 2.1. Intrinsic frame of reference: Ahtna postpositions .............................. 26

Example 2.2. Relative frame of reference: kuzuun ts’en ‘right’ and tl’aghes ts’en

‘left’............................................................................................................... 28

Example 2.3. Affixed directionals .......................................................................... 31

Example 2.4. Directionals in use ............................................................................ 32

Example 2.5. Major River Orientation ................................................................... 35

Example 2.6. Use of ‘upriver’ for points along the Copper River, which is the major

river in the drainage ....................................................................................... 38

Example 2.7. Use of ‘upland’ for points along the Klutina River, which is a tributary

of the Copper River........................................................................................ 39

Example 3.1. Assignment of Tazlina River to diorama........................................... 48

Example 3.2. Use of English north and south in elicitation of Ahtna directionals;

diorama: the Tazlina River ............................................................................. 53

Example 3.3. Nae’ ‘upriver’ only used for north-south rivers; diorama: the

Matanuska River ............................................................................................ 59

Example 3.4. Strength of association between nae’ ‘upriver’ and north on the Tok

and Richardson Highways; diorama: the Copper River................................... 63

Example 3.5. Use of ngge’ ‘upland’ for the upriver direction; diorama: The

Matanuska River drainage .............................................................................. 67

xx

Example 3.6. Tgge’ ‘up (vertically)’ along Tazlina River; diorama: the trail to

Tazlina Lake .................................................................................................. 70

Example 3.7. Confirming tgge’ ‘up (vertically)’; diorama: the trail to Tazlina

Lake............................................................................................................... 71

Example 3.8. Tgge’ ‘up (vertically)’ for true verticals ............................................ 73

Example 3.9. Tgge’ ‘up (vertically)’ on the upper Copper River; diorama: the

Copper River.................................................................................................. 75

Example 4.1. Nse’ ‘ahead’ as ‘coastal’; diorama: the Richardson and Glenn

Highways ....................................................................................................... 89

Example 4.2. Nse’ ‘ahead’ beyond the “hump of the world” toward Anchorage ..... 96

Example 4.3. Nse’ ‘ahead’ beyond the “ball of the world” toward Valdez............ 100

Example 5.1. Direction-describing adverbial verb prefixes in Saen tah xay tah .... 123

Example 5.2. Direction-describing adverbial verb prefixes in Ta’stede dze’ ........ 124

Example 5.3. Direction-describing adverbial verb prefixes in Naaghay................ 125

Example 5.4. Use of directionals in Saen tah xay tah ........................................... 126

Example 5.5. Use of ngge’ ‘upland’ in Naghaay .................................................. 127

Example 5.6. Use of ’utgge’ ‘distantly up vertically’ in Naghaay......................... 127

Example 5.7. Use of ’unaan ‘distantly across’ in Naghaay................................... 128

Example 5.8. Use of yihwits’en ‘from there’ in Ta’stede dze’.............................. 134

Example 5.9. Use of xona ‘then’ in Saen tah xay tah............................................ 138

Example 5.10. Use of relative clauses in Naghaay ............................................... 141

Example 6.1. Use of directionals in discourse ...................................................... 149

xxi

Example 6.2. Ahtna directionals suffixed with postpositions (as opposed to

adverbializers).............................................................................................. 159

Example 6.3. Mr. Sanford’s use of ’u- ‘DIST’ to describe comparatively short

distance ........................................................................................................ 169

Example 6.4. Mr. Sanford’s use of ’u- ‘DIST’ to describe comparatively intermediate

distance ........................................................................................................ 171

Example 6.5. Mr. Sanford’s use of ’u- ‘DIST’ to describe comparatively long

distance ........................................................................................................ 173

Example 6.6. Mrs. John’s use of ’u- ‘DIST’ to describe comparatively short

distance ........................................................................................................ 175

Example 6.7. Mrs. John’s use of ’u- ‘DIST’ to describe a comparatively intermediate

distance ........................................................................................................ 176

Example 6.8. Mrs. John’s use of ’u- ‘DIST’ to describe a comparatively long

distance ........................................................................................................ 178

Example 6.9. Mr. Tansy’s use of ’u- ‘DIST’ to describe comparatively short

distance ........................................................................................................ 180

Example 6.10. Mr. Tansy’s use of ’u- ‘DIST’ to describe comparatively intermediate

distance ........................................................................................................ 182

Example 6.11. Mr. Tansy’s use of ’u- ‘DIST’ to describe comparatively long

distance ........................................................................................................ 183

Example 6.12. Mr. Sanford’s use of na- ‘MED’ .................................................... 187



xxii

Example 6.15. Mr. Sanford’s use of da- ‘PROX’ ................................................... 193

Example 6.16. Mrs. John’s use of da- ‘PROX’ to describe comparatively short

distance ........................................................................................................ 195

Example 6.17. Mrs. John’s use of da- ‘PROX’ to describe comparatively intermediate

distance ........................................................................................................ 196

Example 6.18. Mrs. John’s use of da- ‘PROX’ to describe comparatively long

distance ........................................................................................................ 198

Example 7.1. Literal use of punctual directional suffix -t...................................... 204

Example 7.2. Literal use of area directional suffix -xu.......................................... 204

Example 7.3. Literal use of allative directional suffix -e....................................... 205

Example 7.4. Literal use of ablative directional suffix -dze................................... 206

Example 7.5. Directionals with punctual suffix -t co-occurring with toponyms and

locational nouns ........................................................................................... 208

Example 7.6. Major and minor journey legs and discourse episodes, Mrs. John ... 211

Example 7.7. Major and minor journey legs and discourse episodes, Mr.

Sanford ........................................................................................................ 215

Example 7.8. Use of directionals in discourse ...................................................... 218

Example 7.9. End of major journey leg inititated in (7.8) ..................................... 220

xxiii

LIST OF FIGURES

Figure 1.1. Place of Ahtna in the Dene-Yeniseian Superfamily ................................ 7

Figure 2.1. Scene With Possible Descriptions in Multiple Frames of Reference ..... 25

Figure 2.2. Schematic Representation of the Semantics of Ahtna Directional

Stems ............................................................................................................. 34

Figure 2.3. Schematic Representation of Major River Orientation Principles 1

and 2 .............................................................................................................. 36

Figure 3.1. Fieldwork on Ahtna Directionals; Diorama of the Tazlina River .......... 47

Figure 3.2. Cardinal Directions of the Room in Which the Fieldwork Took

Place .............................................................................................................. 52

Figure 3.3. Elevation Profile of the Glenn Highway ............................................... 58

Figure 3.4. Diorama of the Matanuska River Drainage........................................... 59

Figure 3.5. Diorama of the Copper River, With the Richardson and Tok

Highways ....................................................................................................... 62

Figure 3.6 Diorama of the Trail to Tazlina Lake..................................................... 69

Figure 3.7. Diagram of Semantic Shift ................................................................... 77

Figure 4.1. Fieldnotes Containing References to Relative Distances of nse’ and

nae’................................................................................................................ 85

Figure 4.2. Diorama of the Junction of the Richardson and Glenn Highways ......... 88

Figure 4.3. Elevation Profile (Approximate) of the Glenn Highway ....................... 98

Figure 4.4. Working with Ahtna Region Map......................................................... 99

xxiv

Figure 4.5. Elevation Profile (Approximate) of the Richardson Highway From the

Glenn Highway Junction to Valdez .............................................................. 102

Figure 4.6. Path of the Sun as Seen From Copper Center, December – June......... 105

Figure 7.1. Major and Minor Discourse Episodes in (7.6)..................................... 218

Figure 7.2. Possible Pathway for Language Change Via Discourse ...................... 223

xxv

LIST OF TABLES

Table 1.1. Ahtna Verb Template ............................................................................ 10

Table 1.2. Example of Aspectual Stem Allomorphy in Ahtna: ‘burn’ ..................... 12

Table 1.3. Ahtna Themes Based on Stem yaa ‘SG.go’ and Examples of Fully

Inflected Forms .............................................................................................. 13

Table 2.1. Tripartite Morphemic Structure of Ahtna Directionals........................... 31

Table 4.1. Locations That Are and Are Not in the nse’ ‘ahead’ Directions ............. 83

Table 4.2. Locations by nse’ and Distance.............................................................. 87

Table 4.3. Locations by nse’ and Proximity to Salt Water ...................................... 96

Table 4.4. Locations by nse’ and the Need to Cross a Mountain Pass ................... 103

Table 4.5. Aggregated Possible Meanings of nse’ ................................................ 107

Table 5.1. Ahtna Verb Template .......................................................................... 122

Table 6.1. Tripartite Morphemic Structure of Ahtna Directionals ......................... 150

Table 6.2. Directional Morphology Found in (6.1) ............................................... 146

Table 6.3. Proto-Athabascan Demonstrative Prefixes and Their Ahtna Reflexes as

Directional Prefixes...................................................................................... 157

Table 6.4. Pre-Proto-Athabascan Directional Stems and Their Ahtna Reflexes..... 158

Table 6.5. Proto-Athabascan Adverbializing Suffixes and Their Ahtna Reflexes as

Directional Suffixes ..................................................................................... 161

Table 6.6. Most Frequent Directional Words in the Travel Narrative Data in

Descending Order of Frequency ................................................................... 163

xxvi

Table 6.7. Frequencies of the Directional Prefixes in the Travel Narrative Data in

Descending Order ........................................................................................ 164

Table 6.8. Frequencies of Directional Stems in the Travel Narrative Data in


Table 6.9. Frequencies of Directional Suffixes in the Travel Narrative Data in


Table 6.10. Journey Legs Described by Mr. Sanford Using ’u- ‘DIST’ and Their

Length in Miles ............................................................................................ 168

Table 6.11. Journey Legs Described by Mrs. John Using ’u- ‘DIST’ and Their Length

in Miles........................................................................................................ 174

Table 6.12. Journey Legs Described by Mr. Tansy Using ’u- ‘DIST’ and Their Length

in Miles........................................................................................................ 179

Table 6.13. Distances Described by Mr. Sanford Using na- ‘MED’ and Their Length

in Miles........................................................................................................ 186

Table 6.14. Journey Legs Described by Mrs. John Using da- ‘PROX’ and Their

Length in Miles ............................................................................................ 194

Table 6.15. Summary of Ranges in Miles of Distances Described by Speaker and

Directional Prefix......................................................................................... 198

Table 7.1. Tripartite Morphemic Structure of Ahtna Directionals......................... 203

xxvii

LIST OF MAPS

Map 1.1. Geographic Spread of the Athabascan Family ........................................... 5

Map 1.2. Indigenous Peoples and Languages of Alaska............................................ 8

Map 2.1. The Copper River and Its Tributary, the Klutina River ............................ 37

Map 3.1. The Ahtna Region ................................................................................... 45

Map 6.1. Area Described in (6.1) ......................................................................... 151

Map 6.2. Overview of Travel Routes Described in the Three Narratives .............. 154

Map 6.3 Route Described in Mr. Sanford’s Narrative........................................... 155

Map 6.4. Route Described in Mrs. John’s Narrative ............................................. 155

Map 6.5. Route Described in Mr. Tansy’s Narrative............................................. 156

Map 6.6. Example of ’u- ‘DIST’ for the Route from Ben Tah to the Copper

River ............................................................................................................ 169

Map 6.7. Example of ’u- ‘DIST’ for the Route from Ba’stade i to Natii Caegge.... 171

Map 6.8. Example of ’u- ‘DIST’ for the Route from Saas Dze to Fish Camp Near

Kedi eni Na’................................................................................................. 172

Map 6.9. Example of ’u- ‘DIST’ for the Route from Tes T’aa Menn’ Tes to Taggos

Menn’........................................................................................................... 175

Map 6.10. Example of ’u- ‘DIST’ for the Route from Tsic’e ggodi Na’ to Ts’i ten

Kats’etses Na’ .............................................................................................. 177

Map 6.11. Example of ’u- ‘DIST’ for the Route from Natae de to Copper Glacier and

Back............................................................................................................. 178

xxviii

Map 6.12. Example of ’u- ‘DIST’ for the Route from Tl’ahwdicaax Na’ to Nts’ezi

Na’............................................................................................................... 180

Map 6.13. Example of ’u- ‘DIST’ for the Route from Taben ’aa Tayene’ to Taben ’aa

Bene’............................................................................................................ 182

Map 6.14. Example of ’u- ‘DIST’ for the Route from Taben ’aa Bene’ to Ben

Datgge ......................................................................................................... 183

Map 6.15. Example of na- ‘MED’ for the Area Across from Ts’itae Caegge ........ 187

Map 6.16. Example of na- ‘MED’ for the Relationship Between Ts’itu’ K’et and

Kedi eni Na’................................................................................................. 189

Map 6.17. Example of na- ‘MED’ for the Relationship Between Tsidghaazi Na’ and

Snuu Caegge ................................................................................................ 191

Map 6.18. Example of da- ‘PROX’ for the Route from Una’ Hwnelk’ezi Na’ to

Ts’itae Na’ .................................................................................................. 193

Map 6.19. Example of da- ‘PROX’ for the Route from Kolgiisde to Nearby

Uplands........................................................................................................ 195

Map 6.20. Example of da- ‘PROX’ for the Route from Bes Ce’e to Stl’aa

Caegge......................................................................................................... 196

Map 6.21. Example of da- ‘PROX’ for the Route from Natae de to Tanaade

Menn’........................................................................................................... 197

Map 7.1. Major and Minor Journey Legs in (7.6) ................................................. 214

Map 7.2. Major and Minor Journey Legs in (7.7) ................................................. 217

Map 7.3. Area Described in (7.8) ......................................................................... 219

Map 7.4. Route from Natii Na’ to Ts’itae Na’ ..................................................... 222

1

Chapter 1

Introduction

1.1 Objective of This Dissertation

This objective of this dissertation is to examine one small corner of grammar

embedded in discourse in the Ahtna Athabascan language of south central Alaska:

the use and semantics of the lexical class of directionals. In particular, this

dissertation looks at how Ahtna speakers use directionals in spontaneous discourse

and elicitation against the backdrop of the actual physiography of Ahtna territory.

The semantics of the directional system is traditionally riverine, meaning that the

direction of the flow of the local river local determines which directional term

speakers choose. Talk about direction and location of referents in the natural

landscape is common among Ahtna speakers: Ahtna people are traditionally semi-

nomadic, and verbally displaying one’s knowledge of overland travel through Ahtna

territory has a special place in culture and society (Kari 2010). It serves as an index

of one’s Ahtna identity (e.g. Moore & Tlen 2007). Among the linguistic resources

available to Ahtna speakers for describing concepts like path and location are the

directionals, the use of which is a direct reflection of a speaker’s familiarity with the

geography of the region he or she is describing. Awareness of the local ecology is

thus not only central to Ahtna cultural practices, but also potentially influences the

development of the grammar over time.

2

In this dissertation I am concerned with the relationship between language

change over time and the use of the directionals in discourse and elicitation. The first

major section of the dissertation examines the recent changes in the semantics and

usage of the directionals through the lens of language contact. Using data from my

fieldwork, I show that the nearly constant contact of Ahtna with the dominant

English language is causing a recent, very perceptible shift in the semantics of the

directionals, such that they now refer less to the orientation of the local river, and

more to the cardinal directions found in English.

The second major section of the dissertation looks at language change from a

purely language-internal, usage-based point of view. Using discourse data from a

previous generation of Ahtna speakers, I present evidence that the complex

morphology of directionals is in part lexicalizing, in part grammaticalizing, over

time, leading to a loss of semantic clarity that speakers are compensating for via

other resources in the discourse structure. This study supports claims that discourse –

the way members of a speech community use language for communication – is the

locus of language change, and has implications for assumptions about the

compositionality of polysynthetic words.

Central to the discussion in this dissertation is the topography of the landscape

itself: when studying directional reference, one must have some nonlinguistic data

against which to compare terms that speakers are selecting. To that end, geographic

information systems technology plays a large role in the data presented in the

following chapters.

3

Below I present a general introduction to the dissertation. Section 1.2 discusses

the current linguistic setting of the Ahtna community and provides an overview of

the history of Ahtna language documentation. Section 1.3 presents information on

the genetic affiliation of Ahtna. Section 1.4 is a typological sketch of the language,

designed to give readers unfamiliar with Athabascan grammar enough background to

be able to navigate the data in the following chapters. Section 1.5 reviews the

relevant data on the topic of geographic knowledge and language in Ahtna,

Athabascan, and general linguistics. Section 1.6 is a more specific roadmap of the

following chapters.

1.2 The Linguistic Setting

The Ahtna community today consists of eight modern villages (Mentasta,

Chistochina, Gakona, Gulkana, Tazlina, Copper Center, Chitina, and Cantwell) in

the Copper River and Upper Susitna drainages of south central Alaska. More than

1600 Ahtna residents of the area are shareholders in Ahtna, Incorporated, one of the

thirteen Alaska Native Regional Corporations established by Congress under the

1971 Alaska Native Claims Settlement Act. The Corporation serves a region of more

than 44,000 square miles and is the central body for providing services and cultural

education opportunities to its membership.

Among the Ahtna population, there are approximately thirty first-language

speakers of the language alive today, all of whom are at least sixty years of age.

Although most children know some phrases and vocabulary items in the language,

English is the language taught in homes today. There is, however, a growing passion

4

for Ahtna language revitalization in the community, and several recent projects have

undertaken to increase knowledge of the heritage language among Ahtna people and

non-Ahtna residents of the Copper Valley. Among these include a grant from the

Institute for Museum and Library Services to Ahtna, Incorporated, for the

establishment of a digital archive of Ahtna linguistic and ethnographic recordings; a

grant from the Administration for Native Americans to design language materials for

very young (age 3-5) children; and a regular language class at the Ya Ne Dah Ah

school administered by the Chickaloon Village Traditional Council. Prince William

Sound Community College, the local branch of the University of Alaska system,

occasionally offers Ahtna language classes for adults and young adults in the town of

Glennallen.

Documentation of the Ahtna language goes back as far as the expedition of

Lieutenant Henry Allen, who led a small group of U.S. Army soldiers along the

Copper, Tanana, and Koyukuk Rivers between March and August 1885. His report

to Congress contains detailed geographic and ethnographic information, as well as

some early transcriptions of Ahtna vocabulary, personal names, and toponymy.

The majority of documentation of Ahtna has been conducted by linguist James

Kari. Kari lived in the region for several years in the 1980s and continues to visit

regularly from his current home in Fairbanks. Among his many publications on

Ahtna is a morpheme-based dictionary (Kari 1990) that serves as the most

comprehensive record of lexicon and grammatical notes to date. Kari’s contributions

to the study of Ahtna and Athabascan geographic knowledge, the topic relevant to

this dissertation, are outlined in Section 1.5 below. Importantly, the ethos of Kari’s

5

work with Ahtna speakers has since its inception foreshadowed the principles of

contemporary documentary linguistic practice (e.g., Himmelmann 1998, 2006), in

terms of his dedication to the collection, transcription, and preservation of recordings

of spontaneous language use and his forging of personal relationships with

individual Ahtna speakers.

1.3 Genetic Affiliation

The Athabascan-Eyak-Tlingit language family contains the largest number of

languages and the widest demographic of any family in North America (Mithun

1999). It is also the largest geographically: it encompasses parts of Alaska, Canada,

Washington, Oregon, California and the southwest United States. Within

Athabascan, three major geographic groups occupying noncontiguous areas, are

recognized: Northern Athabascan, Pacific Coast Athabascan, and Apachean

(Southern) Athabascan. Map 1.1 shows the entire Athabascan language area.

Map 1.1: Geographic Spread of the Athabascan Family

(source: Wikipedia)

6

It should be noted that actual language boundaries, especially in Alaska and

Canada, are difficult to establish precisely (Krauss 1973, Krauss & Golla 1981).

Frequent travel and trade caused speaker groups to be in constant contact, resulting

in mutual intelligibility and many shared features (see, for example, Holton 2000 for

a discussion of the difficulty of drawing clear boundaries around the languages of the

Tanana River drainage in Alaska). Exposure to neighboring Athabascan languages

enabled speakers to develop a keen awareness of phonetic correspondences (Krauss

1980), and intermarriage between members of different language communities was

common.

The history of genetic classification of Athabascan-Eyak-Tlingit has been

fraught with controversy. Sapir (1915) posited a genetic relationship between Haida,

Athabascan, and Tlingit (Eyak data was unavailable to him at the time) as

constituting the Na-Dene super-family. His hypothesis was later questioned (Krauss

1964, 1965) based on Sapir’s lack of accurate Haida data. Enrico (2004) revived the

Na-Dene hypothesis as tenable. Recent work by Vajda, Leer, and others (Kari &

Potter 2010) explore the more recent hypothesis establishing a link between

Athabascan-Tlingit-Eyak and the Yeniseian language family of Siberia. The

resulting Dene-Yeniseian superfamily is shown in Figure 1.1; only the subgrouping

from the Athabascan-Tlingit-Eyak node down is considered uncontroversial in the

field. In any case, it is generally accepted among scholars that only geographic,

rather than genetic, subgrouping within Athabascan is possible, owing to conflicting

isoglosses.

7

Figure 1.1: Place of Ahtna in the Dene-Yeniseian Superfamily

As for geographic subgroupings, the Northern group contains the most

languages, with 26 in Alaska and Canada. This group spans a vast land area,

stretching from south central Alaska and the Alaskan Interior on the west, eastward

through the Yukon and Northwest Territories, and south into British Columbia,

Alberta, Saskatchewan and Manitoba. The Southern Alaskan subgroup contains two

languages (Dena’ina and Ahtna), and the Central Alaskan subgroup contains nine

(Deg Xinag, Holikachuk, Koyukon, Hän, Gwich’in, Upper Kuskokwim, Lower

Tanana, Upper Tanana and Tanacross). Map 1.2 shows the indigenous languages of

the state of Alaska (the Athabascan family is indicated in warm colors).

8

Map 1.2: Indigenous Peoples and Languages of Alaska (Krauss et al. 2010)

In Canada, the Northwestern Canadian subgroup contains ten languages

(Tahltan, Tagish, Kaska, Tutchone, Sekani, Beaver, Tsetsaut, Slave, Dogrib and

Chipewyan), and the Central British Columbia subgroup contains five (Babine-

Witsuwit’en, Carrier, Sarcee, Chilcotin and Kwalhioqua-Tlatskani) (Mithun 1999).

The Pacific Coast group contains four languages in Oregon and southern

Washington (Upper Umpqua, Galice, Rogue River and Tolowa), and three in

California (Hupa, Mattole and Eel River). The Apachean group of the southwest

United States is the largest in terms of active speakers: up to 100,000 speakers of

Navajo alone reside in Arizona, New Mexico and Utah. Six other languages also

9

belong to this group (Western Apache, Chiricahua, Mescalero, Jicarilla, Lipan and

Plains Apache) (Krauss & Golla 1981, Mithun 1999).

1.4 Typological Sketch

The aim of this dissertation is not to provide a reference grammar for the

entire Ahtna language, but nonetheless, readers unfamiliar with Athabascan

linguistics will find a typological overview useful. This sketch draws heavily from

the excellent presentations in Cook & Rice (1989) and Mithun (1999). Athabascan

languages are notoriously polysynthetic, fusional, head-marking and head-final.

Verbs are heavily prefixing, and scholars have traditionally used a template model to

describe verb structure; in fact, Athabascan represents the archetypal example of

templatic morphology. The template consists of a stem preceded by up to twenty

prefix slots, the exact schema of which varies from language to language (see, for

instance, Tenenbaum (1978) for Dena’ina, Rice (1989) for Slave, Cook (1984) for

Sarcee, Kari (1979) for Ahtna, Hargus (1988) for Sekani, and Holton (2000) for

Tanacross). Recent work deals with some of the problems inherent in the template

model, and presents alternative models. Rice (2000) proposes a model based on

semantic scope, while McDonough (1999) advocates a bipartite model in which

verbs are compounds, rather than prefixed stems. For ease of discussion, I adhere to

the traditional “slot-and-filler” template model of verb structure in the present study.

The Ahtna verb template is shown in Table 1.1 and is discussed in more detail in

Chapter 5.

10

Table 1.1: Ahtna Verb Template (adapted from Kari 1990)

Bo

un

d p

ost

po

siti

on

al

obje

ct

Der

ivat

ional

/ T

hem

atic

Iter

ativ

e

Dis

trib

uti

ve

Inco

rpora

te

Them

atic

Pro

no

min

al

Qu

alif

iers

Conju

gat

ion

Su

bje

ct

Cla

ssif

ier

Ste

m

11 10 9 8 7 6 5 4 3 2 1

One notable phonological feature of the family is the development of

distinctive tone from Proto-Athabascan constricted syllables (Cook & Rice 1989,

Krauss 2005). Many modern languages do not have tone, including Ahtna and the

entire Pacific Coast group. Those that do, however, have either high or low marked

pitch as a reflex of Proto-Athabascan constriction. Distribution of tone, as Mithun

(1999) notes, is not purely areal: Tanacross is a high tone language surrounded by

low tone languages, and Sekani has low tone in a sea of high tone neighbors.

Kingston (2005) offers a phonetic explanation for the evolution of high and low tone

from tense versus creaky voice qualities, respectively, in constricted syllables.

Athabascan languages have complex aspectual systems, realized through the

interplay of inflectional prefixes, derivational prefixes, and suffixation of the verb

root. Axlerod (1993) recognizes for Koyukon fifteen distinct morphologically

marked aspects, including momentaneous, perambulative, continuative, persistive,

repetitive, bisective, etc., which intersect with four “modes,” or viewpoint aspects:

imperfective, perfective, future and optative. While no one verb can be expressed in

11

every possible combination of aspect and mode, the system allows a great deal of

information to be contained within a single word.

Also of interest is the system of classificatory verbs. Basso (1968) provides a

simple example of the system in Western Apache: nato nt and nato n’a both

mean ‘hand me the tobacco.’ The stem of the former, -t , refers to an elongated

object (i.e., ‘hand me the cigarette’), while the stem of the latter, -’a, refers to a

compact object (i.e., ‘hand me the pack of cigarettes’). Extensive descriptions of the

classificatory verb system can be found in Young & Morgan (1980) for Navajo, Rice

(1989) for Slave, and Jetté & Jones (2000) for Koyukon.

It is important, when dealing with any language family, to understand the

terminology used by scholars in describing the languages. Of particular concern

within the study of Athabascan are the various levels of inflection and derivation

taking place in the process of word building. Historically, “[i]t has long been

recognized in the Athapaskan literature that every major category lexical item ... is

built around a stem where the stem has a constant meaning associated with it” (Cook

& Rice 1989:19, emphasis added). Early literature (e.g. Sapir 1923) considers the

stem to be the lowest level of structure in Athabascan lexical items. Verb stems are

treated as a single entity with multiple allomorphs (collectively called a stem set).

Typical dictionary citations present stem sets as a matrix, with four modes along the

x-axis, and any number of appropriate aspects along the y-axis. Many allophones are

homophonous, and, not surprisingly, not all cells need be filled. Table 1.2 shows an

example stem set from Ahtna.

12

Table 1.2: Example of Aspectual Stem Allomorphy in Ahtna: ‘burn’

Perfective Imperfective Future Optative

MOM -k’aas -k’aan -k’aa -k’aa

DIST -k’an’ -k’aan -k’an’ -k’an’

CONT -k’an’ -k’aan -k’an’ -k’an’

CUST -k’aas -k’aan

NEU -k’an’ -k’aan

NEU-

CUS -k’an’ -k’aan

Later research, however, points towards an even lower level of meaningful

structure than the stem, that of the root (with a stem being composed of a root plus

some categorizing “stem formative” material; in the case of verbs this is a modal

suffix (Kari 1979)). Work by Leer (1979) and Kari (1979) asserts that although roots

are the underlying meaningful form, they can never appear unbound, and Rice (1989)

assumes that these roots do not belong to explicit lexical categories until stem

formation occurs. In other words, roots are the lowest level of meaning, but they are

not manipulable as nouns or verbs until stem affixation has occurred.

Because roots themselves do not belong to lexical categories, they can easily

become verb or nouns by taking the appropriate morphology. Examples of “matched

pairs” of nouns and verbs are abundant in the literature, as in (1.1).

13

(1.1) Noun and verb stems from same root, Slave a. dloh ‘laughter’ -dloh ‘laugh’ b. tth h ‘axe’ -tth h ‘chop with axe’ c. tsih ‘ochre’ -tsile ‘be red’

(Rice 1989:161)

The theme is the basic lexical entry for verbs and is composed of the stem plus

any obligatory thematic prefixes. Table 1.3 shows four themes, or lexical entries,

based on the Ahtna stem yaa ‘SG.go’, along with examples of fully inflected words

based on those themes.

Table 1.3: Ahtna Themes Based on Stem yaa ‘SG.go’ and Examples

of Fully Inflected Forms (Kari 1990:422-424)

Theme (lexical entry)

Gloss of theme Examples of inflected forms

+yaa ‘SG, DUAL goes’ tayaa ‘s/he will come’

daniyaa ‘s/he walked a long time’ naniyaa ‘s/he went across’

c’+ +yaa ‘SG animal goes’ tic’eniyaa ‘it went out into the woods’

nac’idyaa ‘it returned’

cin’#d+l+yaa ‘pretend to walk’ cin’dalyaa ‘s/he is pretending to walk along’

INC# +yaa ‘INC moves’ nata’idya ‘water rose’

neke’o’ayaax ‘bubbles are swirling around’ ka ts’ighiyaa ‘the wind went up’ sc’anaa niyaa ‘I couldn’t sleep (lit. sleep

went away from me)’ ts’inaxay’idyaa ‘winter came back out

again’

14

This section has provided a basic overview of key points in Ahtna and

Athabascan grammar. Chapter 2 discusses in detail the portion of the grammar that is

relevant to this dissertation, the grammar and semantics of the lexical class of

directionals.

1.5 Review of Previous Research on Language and Geography

in Ahtna, Athabascan, and General Linguistics

This dissertation draws on several related fields of linguistic study, including

research on Ahtna and Athabascan geographic knowledge, spatial cognition in

language, and discourse about route description. Major works in each field are

named and briefly described below.

1.5.1 Kari s Work on Geography and Place Names in Ahtna and

Athabascan

Kari has been most prolific on the topic of geographic salience in Northern

Athabascan languages and cultures, and it has been one the primary themes of his

research for three decades. Kari’s contributions in this area can be roughly divided

into four categories: documentation of indigenous toponyms, collections of travel

narratives, analytical discussions for linguistic audiences, and reports for

nonlinguistic audiences. These are discussed below.

15

1.5.1.1 Lists of Toponyms

Dedicated toponym lists for Ahtna are Kari 1983, which contains about 1300

names, and the updated second edition with complete georeferences in Kari 2008,

which contains over 2100 names. A dedicated toponym list for Upper Tanana is Kari

1997. Other place names are documented in Kari & Fall 2003 and Kari 2006 for

Dena’ina, and Kari & Tuttle 2005 for Ahtna.

1.5.1.2 Collections of Travel Narratives

Ahtna travel narratives are found in Kari 1986, 2010. Dena’ina and Ahtna travel

narratives are found in Kari & Fall 2003. Kari (p.c.) notes that it is relevant to the

present study that there are few sources for other languages – Athabascan and

beyond – that parallel the richness of the Ahtna corpus of travel narratives.

1.5.1.3 Analytical Discussions for Linguistic Audiences

Kari’s early analytical work on the topic of Athabascan geographic knowledge

(1989, 1996a, 1996b) discusses how the etymologies of ‘mountain’ and ‘stream’ in

different languages show ancient language group and land use boundaries between

upstream/eastern groups and downstream/western groups. These early papers also

survey the Alaskan Athabascan place name corpora and his own field methodology.

These papers are also the earliest discussions of the systematic principles driving

Athabascan toponymy, which he later refines in Kari (2008, 2010, forthcoming).

These principles are discussed in more detail in Chapter 5.

16

1.5.1.4 Reports for Nonlinguistic Audiences

Linguistic and ethnolinguistic reports for various government agencies relating to

site use, travel, and migration include Kari 1999, Kari & Tuttle 2005, Simeone &

Kari 2005, and Kari 2006.

1.5.2 Other Sources on Na-Dene Geolinguistic Knowledge

Basso (1984, 1988, 1996) discusses the relationship between landscape,

language, and morality among the Western Apache. Cruikshank (1990) is a summary

of linguistic anthropological research on the role of place naming in ethnohistorical

reconstruction among Athabascan speakers in the Yukon. Kelley & Francis (2005) is

a decidedly nonlinguistic discussion of route knowledge and ‘verbal maps’ in Navajo

culture. Holton (2009) compares and contrasts landscape ontology (e.g., terms for

‘hill’) between Eskimo and Athabascan language in regions of Alaska where

speakers share hunting territory. Thornton (2008) is an anthropological examination

of the relationship between place, place naming, site use, and identity among the

Tlingit. Busch (2000) discusses the use of directionals in Gwich’in narrative, which I

describe more thoroughly below. Moore & Tlen (2007) is a sociocultural linguistic

discussion of the use of directionals in indexing a social group’s claims to a

particular tract of land in discourse about territory among Southern Tutchone

Athabascan speakers in the Yukon.

17

1.5.3 Ahtna Ethnography

Anthropological and ethnographic discussions of the Ahtna people, including

land and resource use, are found in Reckord (1979, 1983a, 1983b), de Laguna &

McClellan (1981), Simeone & Kari (2002, 2005), Kari & Tuttle (2005). In particular

among these, Kari & Tuttle (2005) discusses the possible Ahtna cultural practice of

‘caucusing’ on land and geography, and states that all Ahtna speakers are familiar

with 100-200 core place names as a result of this common practice.

1.5.4 Descriptions of Directional Systems in Northern Athabascan

The seminal reference for Athabascan directionals is Leer (1989), which

provides an overview of the historical development of the grammar and semantics of

the systems in Athabascan, Eyak, and Tlingit. This article is discussed in Chapter 6.

The Ahtna directional system is described in Kari (1990, 2008, 2010, forthcoming).

Koyukon directionals are described in Jetté & Jones (2000). Dena’ina directionals

are described in Kari (2007).

1.5.5 Spatial Cognition in Language

Levinson (2003), which is to date the most thorough typological treatment of the

linguistic expression of spatial cognition in language, is most useful here for sorting

out notions of frames of reference, i.e., the reference point upon which a language

bases coordinate systems. Levinson’s three types (absolute, intrinsic, and relative) all

occur in different areas of Ahtna grammar, and are discussed thoroughly in Chapter

2.

18

Another school of thought on language and spatial cognition is exemplified by

Landau & Jackendoff (1993). While this was a seminal article on the topic, it is less

relevant to the present dissertation than Levinson (2003) for two reasons: nearly all

of the observations come from English, and the premise of Landau & Jackendoff

(1993) is that language is an accurate reflection of spatial cognition. We can perceive

tiny fluctuations in topography, but it is yet to be shown that any language, not even

an Athabascan language, is subtle enough to express all details faithfully.

1.5.6 Route Description in Discourse

Experimental studies of the language used in route description have been popular

in psychology and psycholinguistics, but have primarily, or perhaps exclusively,

been undertaken with speakers of European languages. Denis (1997) is a study of

route description students in a Parisian university, and Denis et. al (1999) is a similar

study of residents of Venice. In both cases, subjects were asked to give directions to

a known location as if they were describing the route to someone who was

unfamiliar with the area; the responses were then checked for commonalities in an

attempt to determine which parts of direction-giving are most essential. Although

Ahtna travel narratives have a different pragmatic status from instructional discourse

like giving directions, I believe the structure of the discourse is similar in that in both

cases, the discourse is segmented into paths of movement connected by

‘reorientation points’. Ariane & Denis (2004) examine the usefulness of proper

nouns versus generic nouns (e.g., “Hospital Street” vs. “the hospital”) in giving

19

directions, which may provide an interesting comparison to Ahtna toponyms, which

are coined in a very different fashion from western place names.

Allen (2000) presents three principles of route descriptions: (i) speakers present

the descriptions in correct spatiotemporal order, (ii) speakers concentrate descriptive

information at ‘choice points’ or ‘nodes’, and (iii) speakers use spatial delimiters of

length that are consistent with the principle of mutual knowledge with the area being

described. These are consistent with the Ahtna travel narratives. First, speakers do

name locations in the same spatiotemporal order in which they are encountered in

the landscape. Second, travel narrators do concentrate information at ‘nodes’, i.e.,

named locations (although the pragmatic function of the nodes is to provide

background information about site use, rather than instruct listeners how to proceed).

Third, speakers continually index shared knowledge of landscape and terrain in their

narratives.

1.5.7 Spatial Concepts in Narrative

Very few publications on the role of spatial concepts and directional reference in

narrative structure exist, one of which is fortunately about the Athabascan language

Gwich’in. Busch 2000 is a Master’s Thesis in linguistic anthropology that examines

accessibility of location in a chapter from a Gwich’in travel narrative text. Busch

develops an ‘accessibility hierarchy’ for coding spatial terms in the text, and finds

the following correlations:

20

• Nonaccessible spatial referents are introduced with [DIRECTIONAL+PLACE

NAME], to place the location on the ‘narrative map’ and to set the scene for

the following actions.

• Speakers use [DIRECTIONAL] to refresh a spatial reference that has not been

mentioned for several lines of more or when the focus is on the path of travel.

• Highly accessible spatial referents are referenced with simply

[demonstrative]; these are locations that have very recently been introduced.

A few other marginally relevant sources on spatial reference in narrative

exist. Herman (2001) describes the use of spatial reference in scene-setting in ghost

stories from North Carolina. Payne 1984 is a working paper on strategies for

describing location and spatial relations in Yagua narratives.

1.6 Organization of this dissertation

This dissertation is divided into three major sections. As to the first, Chapters 2-4

discuss the changing semantics of the Ahtna riverine directional system Chapter 2

provides background information on the directionals, including an overview of

Levison’s (2003) spatial frames of reference, the morphology of Ahtna the

directional class, and a description of the principles underlying the semantics of the

riverine system. Chapter 3 presents evidence from my own fieldwork that shows that

a contact-induced shift is taking place in the Ahtna directional system: the riverine

semantics are beginning to resemble the cardinal directions of English, the dominant

language in the community, and are losing their purely riverine basis. Chapter 4 uses

evidence from my fieldwork to further refine the semantics of one particular

21

directional that is traditionally glossed as ‘ahead’, a gloss that only begins to cover

the meaning of this word in Ahtna. Both Chapters 3 and 4 present evidence that the

Ahtna directional system differs from the English cardinal system in another crucial

way: rather than defining directions in two dimensions, the Ahtna system also takes

elevation into account, which makes directional reference a three-dimensional

system.

The second section of the dissertation, found in Chapter 5, is a discussion of the

prevalence of directional reference in Ahtna travel narration as compared to that

found in ‘frog-story’ narratives, a common elicitation tool used in linguistic typology

for revealing how a given language refers to path and location. This chapter finds

that frog-story narratives are lacking in the rich description of path that the language

is capable of encoding, and that a description of Ahtna’s path-describing resources

should come from the indigenous genre of travel narration rather than from the

artificial task of telling a frog story.

The third section discusses language-internal change in the directional

morphology. Using spontaneous travel-narration data from a previous generation of

Ahtna speakers (Kari 2010), I show that contrary to linguists’ claims about the

extreme compositionality of the directional morphology, speakers are in fact limiting

themselves to a small number of directional forms even when describing regions of

great physiographic diversity. Using geographic information systems technology, in

Chapter 6 I show that the deictic distinction of near vs. intermediate vs. far distances

that were previously encoded by the directional prefixes is bleaching, and that these

prefixes no longer have full semantic richness they may have had in previous times.

22

The prefixes are lexicalizing to the stem. Chapter 7 examines the suffixes, and shows

that these are grammaticalizing, taking on the function of providing structure to the

discourse and recapturing some of the deictic distinctions that have been lost in the

prefixes. Chapter 8 contains concluding remarks. The Appendix serves as a guide to

the reader for decoding the Ahtna examples contained herein, with notes on

transcription conventions and information on transcription and glossing.

23

Chapter 2

Background on Directionals

This chapter provides background information on the morphology and semantics

of the Ahtna riverine directional system. Section 2.1 places Ahtna within Levinson’s

(2003) typology of spatial frames of reference; Section 2.2 describes the morphology

of the lexical class of directionals; Section 2.3 describes the semantics of the largely

riverine directional stems; and Section 2.4 describes the principle of “local major

river orientation,” which guides the assignment of the directional system in a given

location such that (i) minor streams and tributaries adopt the absolute directions of

their major river, and (ii) speakers shift to different absolute directions when

referring to locations in different drainages.

2.1 Spatial Frames of Reference in Ahtna Grammar and

Lexicon

Levinson 2003, which is to date the most thorough typological treatment of the

linguistic expression of spatial cognition in language, is useful here for sorting out

notions of frames of reference, i.e., the reference points upon which linguistic

coordinate systems are based. Levinson describes three types of frames of reference,

intrinsic, relative, and absolute. All three frames necessarily involve a figure, which

Levinson defines as “the object to be located;” and a ground, “the object with respect

to which the figure is to be located” (2003:41).

24

The first of these frames is the intrinsic frame of reference. This frame expresses

spatial relationships based on some facet of the ground object. For example, in the

English statement the tree is in front of the house, the spatial relationship between

the figure object (the tree) and the ground object (the house) is described in terms of

an intrinsic part of the ground object, i.e., the house’s front.

The second frame of reference is the relative frame, in which spatial

relationships are described based on some facet of the viewer. For example, in the

English statement the tree is to the left of the house, the relationship is described

based in terms of the right and left side of the viewer.1

The third frame of reference is the absolute frame, in which spatial relationships

are determined by some feature of the larger environment. The English cardinal

directional system is an absolute frame of reference: the concept of north in the tree

is to the north of the house is based on the larger environment, rather than some facet

of either the figure object or the ground object.

As Levinson is careful to point out, a language can contain multiple frames of

reference: “I can happily say of [an] assemblage (ego looking at a car from the side,

with the car’s front to ego’s left [as in Figure 2.1]): ‘The ball is in front of the car’

and ‘The ball is to the left of the car’, without thinking that the ball has changed its

place” (2003:25, emphasis added). Similarly, all three types of frames of reference

1 The English relative frame of reference is based on the right and left side of the

viewer under unmarked circumstances. It is possible to use right and left to refer to these facets in the ground object, but this is a pragmatically marked situation and requires further exposition, e.g. the tree is to the left of the house—I mean the house’s left!

25

occur in different areas of Ahtna grammar. These are described in Sections 2.1.1

through 2.1.3 below.

Figure 2.1: Scene With Possible Descriptions

in Multiple Frames of Reference (after Levinson 2003:25, Fig. 2.1)

Levinson (2003) also provides a series of “rotation tests”, which are designed to

test the properties of linguistic spatial constructions and help determine which frame

of reference a particular linguistic system belongs to. The rotation tests involve an

imaginary array of a figure object and a ground object (in Levinson’s examples, a

ball and a chair respectively), as well as an imaginary viewer. These items are then

rotated in various combinations, and the linguistic description of the spatial

relationship is then checked to see if it would still hold after rotation. Each frame of

reference displays a different pattern of spatial properties under rotation.

In an intrinsic frame of reference, if the ball is described as being in front of the

chair, and the figure-ground array is rotated, the ball is still in front of the chair. If

only the ground object is rotated, the ball is no longer in front of the chair. If only a

viewer observing the scene is rotated, the ball is still in front of the chair. In other

words, in a rotation test of an intrinsic frame of reference, a spatial description will

26

hold for a rotation of (i) the viewer or (ii) the figure-ground array, but will fail for a

rotation of (iii) the ground object.

In a relative frame of reference, if the ball is to the left of the chair and the

figure-ground array is rotated, the ball is no longer to the left of the chair. If only the

ground object is rotated, the ball is still to the left of the chair. If only the viewer is

rotated, the ball is no longer to the left of the chair. In other words, in a rotation test

of a relative frame of reference, a spatial description will hold for a rotation of (i) the

ground object, but will fail for a rotation of (ii) the viewer and (iii) the figure-ground

array.

Finally, in an absolute frame of reference, if the ball is described as to the north

of the chair and the figure-ground array is rotated, the ball is no longer to the north

of the chair. If only the ground object is rotated, the ball is still to the north of the

chair. If only the viewer is rotated, the ball is still to the north of the chair. In other

words, in a rotation test of an absolute frame of reference, a spatial description will

hold for a rotation of (i) the viewer or (ii) the ground object, but will fail for a

rotation of (iii) the figure-ground array.

2.1.1 Ahtna Intrinsic Frame of Reference

The Ahtna postpositional system displays an intrinsic frame of reference.

Examples are given in (2.1).

(2.1) Intrinsic frame of reference: Ahtna postpositions

(a) 01 MP; ’A cesi ne-,

handbag/war.bag HES

27

02 c’ecele ’a cesi yii ’iyaas 3S.INDEF.POSS.younger.brother handbag/war.bag in 3S.SUB.go.IMPERF

den, SP/TMP.ADVZ

‘When she went in the youngest brother’s war bag […]’

((Markle Pete, oai:paradisec.org.au:ALB01-041, 00:15:11.720-00:15:15.900))

(b) MP; Deniigi uts’e’ nghedzen.

moose 3S.OBJ.to 3S.SUB.stand.IMPERF

‘The moose is standing to (i.e., facing) her.’


(c) 01 WH; Jesus, (0.6) 02 nen’ k’e tghit’a koht’aen ’e

earth on 3S.SUB.clf.cpt.obj.PERF people CONJ

nakalnic. 3S.SUB.tell.PERF

‘Jesus was on earth; he told people stories.’

((Walya Hobson, oai:paradisec.org.au:ALB01-074, 00:00:00.810-00:00:03.810))

In example (2.1) above, for instance, deniigi ‘moose’ is the figure, and the 3rd

person singular object pronoun u- ‘him/her/it’ is the ground. Their spatial

relationship is defined by the postposition ts’e’ ‘to; facing” (i.e., to the girl’s front).

The frame of reference here is intrinsic, based on the ‘front’ facet of the girl.

This relationship passes Levinson’s three rotation tests for intrinsic frames of

reference. First, if the entire array of figure and ground – here, moose and girl – is

rotated, the spatial relationship of uts’e’ ‘to; facing her’ is maintained. Second, if

only the ground object – the girl – is rotated, the relationship is no longer described

as uts’e’ ‘facing her’ (it is instead ughanii ‘behind her’). Third, if a viewer observing

28

the array is rotated, the relationship between the moose and the girl is still uts’e’ ‘to;

facing her’.

2.1.2 Ahtna Relative Frame of Reference

Ahtna has terms for ‘left’ (tl’aghes ts’en) and ‘right’ (kuzuun ts’en) that

display a relative frame of reference. Kari notes that relative terms do not occur in

travel narratives (2010:129), and in (2.2) speaker Markle Pete confirms that kuzuun

ts’en and tl’aghes ts’en would only be used in a geographic sense under unusual

circumstances like giving driving directions offsite to someone who is not familiar

with the area in question. In (2.2), I had just asked Mr. Pete if he would use kuzuun

ts’en and tl’aghes ts’en to describe turning toward Valdez or Fairbanks upon arrival

at the junction of the Glenn and Richardson Highways (the “Y” in line 11). Mr. Pete

is clear that one would use those terms only for describing the junction to someone

unfamiliar with the region (who presumably would not know the local watershed)

from a location at which neither the junction nor the surrounding topography could

be seen.

(2.2) Relative frame of reference: kuzuun ts’en ‘right’ and tl’aghes ts’en ‘left’

01 MP; Uh-uh, 02 AB; [You would never say that?,] 03 MP; [we don’t use] that too much. 04 AB; No?, not too much. 05 MP; No.

(0.5) 06 AB; Not for driving and stuff.

(2.0) 07 MP; Unless you were,

(1.0)

29

08 stay home, 09 and you:, 10 tell them where to go, 11 when you get to the Y?, 12 AB; Mm-hm?,

(1.6) 13 MP; kuzuun-,

ABAN

14 kuz-, ABAN

(0.5) 15 kuzuun ts’en,

right.side

16 you go Valdez?, (0.9)

17 ehn, HES

(1.4) 18 ehn,

HES

19 tl’aghes ts’en, left.side

20 you go Fairbanks. ‘To the right side, you go to Fairbanks; to the left side, you go to Valdez.

(0.7) 21 AB; Okay.

(0.4) 22 So you would say that.

(0.2) 23 Only at a, 24 MP; If, 25 if you stay home, 26 if he don’t know where he going.


2.1.3 Ahtna Absolute Frame of Reference

The Ahtna directional system is an absolute system, meaning that orientation is

determined by a feature of the larger environment; in this case, that larger feature is

the direction of the flow of the local river. The directional system passes Levinson’s

30

three rotation tests for absolute systems: if a man is standing downstream (daa’) of a

house and we (i) rotate the viewer 180 degrees, the man is still downstream of the

house; (ii) rotate the house 180 degrees, the man is still downstream of the house;

and (iii) rotate the entire array of house and man, the man is no longer downstream

of the house. It is the absolute directional system that is focus of this and the coming

chapters.

2.2 Morphology of the Directionals

Like other Athabascan languages, Ahtna has a separate lexical class of

directional words. Kari (1985, 1990, 2008, 2010) analyzes the Ahtna directionals as

having a tripartite morphemic structure: a stem expressing orientation (a system that

is largely, but not completely, riverine), an optional prefix expressing relative

distance or concepts like ‘straight’, ‘adjacent to’, etc., and an optional suffix that

expresses either a punctual vs. areal distinction or an allative vs. ablative distinction

(see Kari 1985, 1990, 2008, 2010; Leer 1989; Moore & Tlen 2007) This is shown in

Table 2.1.

31

Table 2.1: Tripartite Morphemic Structure of Ahtna Directionals

(from Kari 1990)

Prefixes Stems Suffixes

da- PROX nae’ ‘upriver, behind’ -e ALL

na- MED daa’ ‘downriver’ -dze ABL

’u- DIST ngge’ ‘from water, upland’ -t PUNC

ts’i- ‘straight, directly’ tsen ‘toward water, lowland’ -xu AREA

ka- ‘adjacent’ naan ‘across’

P+gha- ‘from P’ tgge’ ‘up vertically’

n- ‘neutral’ igge’, yax ‘down vertically’

hw- AREA ’an ‘away, off’

nse’ ‘ahead’

Directionals can appear either as unaffixed stems, or as affixed forms – with

considerable morphophonemics,2 as is typical of Athabascan polysynthesis.

3

Examples of affixed directionals are given in (2.3).

(2.3) Affixed directionals

’udaa’a ’utsene DIST.downriver.ALL DIST.downland.ALL

‘to distantly downriver’ ‘to distantly downland’

nanii nanaa MED.upriver.PUNC MED.across

‘intermediately upriver’ ‘intermediately across’

’unggat niidze DIST.upland.PUNC upriver.ABL

‘a point distantly upland’ ‘from upriver’

2 Because of the fusional nature of polysynthesis in Ahtna, I do not give

morphemic parses for directionals throughout this dissertation. Furthermore, it would be cumbersome to provide a list of all possible directional forms. Therefore the reader may refer to the glosses to interpret the morphology of the directionals, which will consistently give glosses for all morphemes present, even when it is not possible to indicate morpheme boundaries.

3 Kari splits the directionals into two groups, calling the bound forms the true

directional class, and the bare stems the “adverbials” (1990:633). Because of shared morphological behavior and the fact that all directionals perform adverbial functions, I do not make this distinction.

32

’unuuxe ’utsiit DIST.upriver.AREA DIST.downland.PUNC

‘a general area distantly upriver’ ‘a point distantly downland’

’uyggu dan’e DIST.down.AREA upriver.ALL

‘a general area distantly vertically down’ ‘to near upriver’

nanggu’ dunse’ MED.upland.AREA PROX.ahead

‘a general area intermediately upland’ ‘near ahead’

An example of the use of directionals in spontaneous narrative is shown in (2.4).

Note that the speaker uses three variations built on the stem ngge’ ‘upland’.

(2.4) Directionals in use

01 KJ; Yihwts’en xona danggeh ta, DEM.AREA.from then PROX.upland.ALL among

(1.6) 02 c’ena’,

stream

(0.9) 03 ngge’ ta kets’ede dze’

upland among 1PL.SUB.pl.go.IMPERF.against.a.place thus

yi ya xungge’, there ADJACENT.upland

(1.6) 04 kecdilaa de.

they.have.names SP/TMP.ADVZ

‘From there in the uplands, as we go on upland, and there are names in the upland there.’

((Katie John, Natae Nenn’ ‘The Batzulnetas Country’, 07:21.186–07:31.430. Kari 2010:84))

2.3 Semantics of the Directional Stems

For this and the next two chapters, I am concerned primarily with the semantics

of the stems of the directional system, as opposed to that of the affixed forms. The

directional system is largely, though not completely, riverine; that is, the frame of

33

reference of the system is based on the orientation of the flow of a river of some

cultural and cognitive importance (more on this below; see Kari 1985, 1990, 2008,

2010; Leer 1989). Kari (2010:130) points out that the stems exist in orthogonally

oppositional pairs (across::across (i.e., in both directions), upriver::downriver,

upland::downland, up (vertically)::down (vertically); see also Leer 1989 for Na-

Dene). The directionals can also be used to describe indoor locations and for some

objects with intrinsic fronts and backs, like large game animals, rifles, etc.4 Figure

2.2 shows a schematic representation of the stems given in Table 2.1 above.

4 Traditional Ahtna houses were built with the front door facing the river, which

allows the interior space to tie in immediately with the directional system. How directionals are used indoors today, when many houses do not face the river, is a topic for future inquiry. See Kari (2010:130-132) or Leer (1989) for more information.

34

Figure 2.2: Schematic Representation

of the Semantics of Ahtna Directional Stems

2.4 “Major River Orientation, Principles 1 and 2” of the Axis of

the Riverine Spatial Grid

Two principles about the orientation of the axis of the Ahtna (and northern

Athabascan) riverine spatial grid have been noted in the literature; I am terming them

“Major River Orientation, Principles 1 and 2”. The principles are shown in (2.5):

35

(2.5) Major River Orientation:

Principle 1: The major axis of the Ahtna directional spatial grid aligns with the major river in the particular drainage about which they are speaking. Principle 2: Ahtna speakers ascribe to minor streams and tributaries the spatial grid of the major river.

In other words, speakers will shift their mental coordinate system when crossing

from one drainage to another to match the flow of the major river in the new

drainage (Major River Orientation Principle 1; Leer 1989, Busch 2000, Levinson

2003, Kari 2010), and all minor rivers in a particular drainage adopt the directional

system of the major river they feed into (Major River Orientation Principle 2; Kari

2008, 2010). These principles are illustrated in Figure 2.3, which shows a schematic

representation of two adjacent river drainages separated by a mountain range. Thick

lines indicate the major river, while thin lines represent tributaries flowing into the

major river.

In Drainage 1, location A is considered to be nae’ ‘upriver’ of location B, but

location C, while indeed upriver of location D along the tributary, is in fact

considered in Ahtna to be ngge’ ‘upland’ of D. Ngge’ refers to locations that are in a

roughly perpendicular direction from the flow of the major river, which includes

points along the tributaries. This illustrates Major River Orientation Principle 2,

which requires Ahtna speakers to have an awareness of the local ecology so that they

know which waterways are tributaries and which are major rivers. Note how

Principle 2 contrasts with English upriver and downriver; an English speaker is

36

likely to refer to location C as being upriver (or perhaps upstream) of location D,

based on their placement along a waterway of any size.

Figure 2.3: Schematic Representation

of Major River Orientation Principles 1 and 2

Crossing from Drainage 1 into Drainage 2 brings a shift of the absolute system to

match the current of the major river in the new drainage. Consistent with the flow of

the major river in Drainage 2, location A is considered to be nae’ ‘upriver’ of

location B , even though the relationship between these two locations is cardinally

perpendicular to the relationship between A and B. Likewise, C is ngge’ ‘upland’ of

D . This illustrates Major River Orientation Principle 1. Note how Principle 1

37

contrasts with the English absolute frame of reference based on cardinal directions,

in which the relationship between locations C and D would be described with the

same term (e.g., north or south) as that used to describe the relationship between

locations A and B .

Kari (2008) gives a striking example of Major River Orientation Principle 2 from

a 49 minute recording by Chief Jim McKinley from 1981 about the geography of the

Copper River region. In particular, Mr. McKinley discusses locations found on the

Copper River and its tributary, the Klutina River, shown in Map 2.1

Map 2.1: The Copper River and Its Tributary, the Klutina River

Kari writes,

It is especially interesting to see the automatic way in which expert Ahtna speakers contrast the main drainage with the large tributaries when they list place

38

in narrative. During [a] recording session with Chief Jim McKinley in January of 1981 Jim summarized with brilliant detail 70 major Ahtna village sites along the Copper River for 27:47 min. Next in a separate segment he mentioned 48 place names going up the Klutina River for about 17 min. […] contrasting ‘upstream’ vs. ‘upland’. (2008:23)

Two excerpts from the recording of Chief McKinley are below. Note in (2.6) his

use of forms of nae’ ‘upriver’ when referring to traveling upriver on the major

Copper River, and in (2.7) his use of forms of ngge’ ‘upland’ when referring to

traveling upstream on the Klutina River. Tl’aticae’e ‘rear water mouth’ from IU 01

of (2.7) is the location of the village of Copper Center, which is shown on Map 2.1 at

the junction of the two rivers.

(2.6) Use of ‘upriver’ for points along the Copper River, which is the major river in

the drainage

01 JM; Yet kanii du’, DEM ADJACENT.upriver.PUNC FOC

(1.3) 02 yet kanii ye du’,

DEM ADJACENT.upriver.PUNC then

(0.3) 03 Ggux,

worm

(0.2) 04 Hwts’iniyaaden see.

3S.SUB.sg.go.PERF.AREA.from.SP/TMP.ADVZ

‘From there the next place upstream is ‘where a monster emerged’.’

(1.0) 05 Ggux,

worm

(0.2) 06 Hwts’iniyaaden,

3S.SUB.sg.go.PERF.AREA.from.SP/TMP.ADVZ

(0.2)

39

07 dae’ hwdi’aan see. thus 3SG.has.name.IMPERF.AREA

‘It is called ‘where a monster emerged’.’

[24 IUs about Ggux Hwts’iniyaaden omitted]

08 Yet kanaa ’u-, DEM ADJACENT.across ABAN

(0.3) 09 ye kanaa ’ud-,

DEM ADJACENT.across ABAN

(0.7) 10 kanaa ’unii yak’a xona,

ADJACENT.across DIST.upriver.PUNC still then

(0.7) 11 Tsedi Kulaen den dae’

copper 3S.SUB.exist.IMPERF SP/TMP.ADVZ thus

hwdi’aan see. 3SG.has.name.IMPERF.AREA

‘There on the other side and upstream is called ‘where copper exists’.’

[17 IUs about Tsedi Kulaen omitted]

12 C’u kanii yak’a, adjacent.upriver.PUNC still

(0.8) 13 T’aghes Ciit,

cottonwood peninsula

(1.4) 14 T’aghes Ciit dae’ hwdi’aan.

cottonwood peninsula thus 3SG.has.name.IMPERF.AREA

‘The very next place upstream there is named ‘cottonwood point’.’

((Jim McKinley, ’Atna’ K’et Kayax ‘Ahtna Villages on the Copper River’, 00:08:51.324-00:10:09.402. Kari 2010:11))

(2.7) Use of ‘upland’ for points along the Klutina River, which is a tributary of the

Copper River

01 JM; Yii ucae’e yegha 3S.NH 3S.POSS.river.mouth 3S.OBJ.in.relation.to

ts’ini ’aayi gha su 3S.SUB.linear.extends.IMPERF.straight.REL in.relation.to EMPH

Tl’aticae’e, rear.water.mouth

40

(0.9) 02 dae’ konii de.

thus 3PL.SUB.say.IMPERF SP/TMP.ADVZ

‘There at the mouth the current flows by there, thus it is said ‘rear water mouth’ [Copper Center].’

(1.3) 03 Ye kanggat,

DEM ADJACENT.upland.PUNC

(1.2) 04 ye du’,

then

(0.6) 05 Ts’ekul’uu’i Cae’e dae’

one.that.washes.out INDEF.POSS.river.mouth thus

konii. 3PL.SUB.say.IMPERF

‘The next place upland of there then is ‘one that washes out mouth’, thus it is said.’

[14 IUs about Ts’ekul’uu’i Cae’e omitted]

06 Ye kanggat du’, DEM ADJACENT.upland.PUNC FOC

(1.1) 07 Ba’ane Ts’ilaaggen Tak’adze’ dae’,

outside someone.killed.him spring.POSS thus

‘The the next place upland is ‘spring of someone killed him outside’.’

[33 IUs about Ba’ane Ts’ilaaggen Tak’adze’ omitted]

08 Yet kangga, DEM ADJACENT.upland

(0.3) 09 ya’a du’,

FOC

(2.9) 10 Tak’ats’ Kaghi ’aaden dae’,

spring linear.extends.IMPERF.up.SP/TEMP.ADVZ thus

‘Next upland there is ‘spring water flows up’ thus.’

((Jim McKinley, Tl’atina’ Ngge’ ‘The Klutina River Drainage’, 00:04:01.660-00:05:25.050. Kari 2010:27-28))

41

2.5 Conclusion

This chapter has provided a review of background information on Ahtna

directionals. Section 2.1 discussed the placement of Ahtna within Levison’s (2003)

typology of frames of reference, showing that his three frames of reference –

intrinsic, relative, and absolute – all occur in different areas of Ahtna grammar: the

postpositional system, the terms tl’aghes ts’en ‘left’ and kuzuun ts’en ‘right’, and the

riverine directional system, respectively. Section 2.2 described the tripartite

morphology of the directionals: a semantically riverine stem, with an optional deictic

or postpositional prefix and an optional suffix expressing a punctual/areal distinction

or an allative/ablative distinction. Section 2.2 also contained examples of affixed

directionals, both in isolation and in spontaneous speech. Section 2.3 described the

riverine semantics of the directional stems. Section 2.4 discussed two principles from

the literature about how the axis of the riverine spatial grid is determined for Ahtna,

which together I am calling Major River Orientation, Principles 1 and 2. Principle 1

states that the major axis of the Ahtna directional spatial grid aligns with the major

river in the particular drainage about which they are speaking. Principle 2 states that

Ahtna speakers ascribe to minor streams and tributaries the spatial grid of their major

river. This section also presents an example of Principle 2 from spontaneous Ahtna

speech.

Looking ahead, this chapter lays the groundwork for the following two chapters,

both of which are based on my fieldwork. In Chapter 3, I discuss how the principle

of Major River Orientation is undergoing semantic shift as a result of contact with

42

English. In Chapter 4, I present some further refinements for the semantics of the

riverine stem nse’ ‘ahead’.

43

Chapter 3

Contact-induced Semantic Change

in Ahtna Directionals

This chapter presents evidence, based on a close analysis of recordings from my

field research with an Ahtna speaking consultant, for a newly occurring change in

the semantic basis of the Ahtna riverine directional system. This change is being

brought about by constant interaction by speakers of Ahtna with speakers of English,

to the effect that the Ahtna directional system is beginning, in some small but

meaningful ways, to resemble the English cardinal system. This change is presented

in terms of my own interactions with a speaker of Ahtna on the subject of

directionals; in many ways our teacher-student interactions mirror the interactions

my consultant has with non-Ahtna speakers in his community every day.

Section 3.1 describes the relevant geography of the Ahtna region and presents a

reference map. Section 3.2 describes the field-based research program on the current

status of Ahtna directionals, and section 3.3 presents the evidence that led me to

form the hypothesis that the Ahtna directional system is undergoing contact-induced

semantic change. Section 3.4 presents crucial evidence that the principles of Major

River Orientation are shifting in Mr. Pete’s conception of the directional system; in

particular, that nae’ ‘upriver’ is becoming tied to the English notion of north. Section

3.5 presents some of the strategies that Mr. Pete uses to compensate for the gap left

in the Ahtna paradigm when nae’ is no longer appropriate for ‘upriver’ in all cases.

44

3.1 The Ahtna Region

The topic of semantic change discussed in this chapter requires the reader to have

some familiarity with the geography of the Ahtna region, including the location of

towns, villages, highways, and major bodies of water. The traditional Ahtna

language area covers more than 35,000 square miles of south central Alaska,

consisting mostly of boreal forest and permafrost, and encompassing parts of three

mountain ranges (the Alaska Range, the Chugach Mountains, and the Wrangell

Mountains) and three major river drainages (the Copper, Matanuska, and Susitna

Rivers) with countless tributaries, lakes, and glaciers. Most of the territory is

unpopulated wilderness, with the majority of the region’s nearly 4000 residents

centered in a few small towns and Native villages along several two-lane highways

connecting to Anchorage to the southwest, Fairbanks to the northwest, and Valdez to

the south. The region contains only a few other roads. See Map 3.1.

Map 3.1 (overleaf): The Ahtna Region

46

3.2 The Bilingual Fieldwork Conditions

The data presented in this chapter and the next come from my fieldwork with

Markle Pete. Mr. Pete is an ideal consultant for studying Ahtna directional reference

because of his knowledge of both the language and the geography of the region. He

is a first-language speaker of Ahtna who is also fluent in English, and he has lived in

the Copper River valley his entire life. In his youth he traveled on foot at least as far

as Mentasta to the north (Map 3.1, E7), the flatlands between the Copper River and

the Wrangell Mountains to the east (Map 3.1, F-G6), and to Tazlina Lake and the

Louise/Crosswind/Ewan Lakes region to the west (Map 3.1, F-G4). As an adult Mr.

Pete worked in construction, building major sections of the Richardson, Tok, Glenn,

and Edgerton Highways, as well parts of the Trans-Alaska Pipeline. Even today he

drives several times each month to Anchorage and occasionally to Fairbanks and

Valdez. Mr. Pete is also currently employed as a teacher at the Ya Ne Dah Ah

School (Map 3.1, G2), which is administered by the Chickaloon Village Traditional

Council and is one of a handful of Alaska Native elementary schools. He is a

respected tribal Elder and culture bearer for the Ahtna people, and is considered in

his community to be an especially patient and engaging teacher of Ahtna language

and tradition. He is also a modest teacher, one who will admit when he does not

know the answer to a question.

My work with Mr. Pete on the topic of directional reference took place during

four videotaped sessions over one week in early March 2010. The first three sessions

47

(oai:paradisec.org.au:ALB01-0571, oai:paradisec.org.au:ALB01-059,

oai:paradisec.org.au:ALB01-060) involved the use of tabletop dioramas as stimuli,

and the fourth (oai:paradisec.org.au:ALB01-062) used a printed Ahtna-language

map of the region as a stimulus.

The tabletop dioramas were made of colored fabric representing different

landforms (e.g., snow-covered earth, grass-covered earth, bodies of water, highways)

and figurines of people, cars, rifles, mountains, trees, and appropriate animals like

moose, foxes, dogs and tree squirrels. See Figure 3.1

Figure 3.1: Fieldwork on Ahtna Directionals; Diorama of the Tazlina River.

Pictured: the author (left) and Markle Pete (right).

During the time in which this research took place, I had been conducting

linguistic fieldwork in the Copper River valley for about seven months over the

course of three years. Throughout that period Mr. Pete was my primary consultant,

and we had been meeting between three and five times each week to study the Ahtna

1 Throughout this dissertation references of this form are permanent handles that

refer to items contained in the author’s archived collection of digital field materials in the Pacific and Regional Archives for Digital Sources in Endangered Cultures.

48

language. Thus the data that is presented here occurred after Mr. Pete and I had built

a friendship and a trusting student-teacher relationship, which is reflected in the

direct and sometimes joking tone of our interactions. It must also be noted that Mr.

Pete is the only Ahtna consultant with whom I formed a close enough relationship to

do the video-based research contained in this chapter and the next. Mr. Pete told me

he considered me to be a good student of Ahtna; nonetheless it should not be

forgotten that I was a guest in the Ahtna community, and many of the conversations

presented here show Mr. Pete working hard to accommodate my lack of knowledge

of the Ahtna language and the worldview presented in the directional system.

At the outset of the research program on the semantics of Ahtna directionals, my

intention was to use the dioramas to create a fictional, anonymous landscape for the

purpose of eliciting directionals.2 However, moments after I laid down white and

blue cloths representing a snow-covered landscape and a river, Mr. Pete immediately

assigned to the blue fabric the identity of the Tazlina River, which is the closest

tributary of the Copper River to Mr. Pete’s home outside of the village of Tazlina.

This exchange is shown in (3.1); note that elicitation sessions, like most of daily life

for Mr. Pete, are bilingual.

(3.1) Assignment of Tazlina River to diorama 01 AB; Ga du’ na’ ((sic; intended: c’ena’ ‘INDEF.POSS.river’)),

DEM FOC intended:river

(0.4) 02 caek’e.

river.mouth

2 It was never my intent, however, to conduct a comprehensive research program

on spatial cognition and language in Ahtna, i.e., the series of experiments presented in Levinson (2003).

{AB indicates river mouth}

49

‘This is a river; a river mouth.’

(2.0) 03 Alright?

(0.3) 04 What [did I?,] 05 MP; [U]na’?,

3S.POSS.river

‘A river?’

(0.5) 06 AB; Uh-[huh]? 07 MP; [That’s] r-, 08 a river, 09 AB; Una’,

3S.POSS.river

10 yeah, 11 uh-huh?.

(0.4) 12 Aen’,

yes

13 una’, 3S.POSS.river

‘A river, yes, a river.’

(0.2) 14 [Okay], 15 MP; [Taz]lina’,

Tazlina.River

‘The Tazlina River.’

(0.3) 16 AB; Okay. 17 @@@@.


Prior to this exchange, I had not told Mr. Pete the nature or the purpose of the

day’s activity. In IU 01 I explain the identity of the blue cloth, attempting to keep the

river and the river mouth fictional by not using the third person possessive prefix u-

(which I did incorrectly, the form for river in IU 01 should have been c’ena’

‘INDEF.POSS.river’). In IU 04 I start to ask Mr. Pete what did I say, a question he

understood before I finished. In IUs 05 and 07-08 Mr. Pete confirms that he

{MP gestures downriver}

{AB gestures downriver}

{MP gestures downriver}

50

understands that the fabric represents a river, although not an indefinite

‘unpossessed’ river, because he uses the 3S possessive prefix u- in IU 05. In his very

next turn (IU 15) he names it as the Tazlina River although I had given him no

reason to do so. Mr. Pete’s rejection of an imaginary river and a fictional landscape

leads me to assume such things are perhaps less important to him than thinking and

talking about real locations. This was interesting to me, because some of the

previous literature on the semantics of Athabascan directionals (e.g. Leer 1989; Kari,

1989, 1996a, 1996b) presents the systems as abstract and widely applicable to any

landscape one may encounter. As we shall see, however, to Mr. Pete the system is

far from abstract, and is instead grounded in the very physiography of the landscape

and the communicative habits of the speech community. As with any part of

grammar, the usage of the Ahtna directional system by speakers on a daily basis

influences the development of the semantics and structure of that system. The

landscape of Ahtna territory influences how Ahtna people used the directional

system to talk about it, and over time with repeated use across the speech

community, the directional system has shifted from a proto-Athabascan system like

that presented in Leer (1989) and discussed in Chapter 6 to one with region-specific

semantics that are implicitly agreed upon by the speech community. Once the

demography of the speech community changes, as it has been doing for the last

century via intensive contact with English and a rapid decline in the number of

Ahtna people who speak the language, the directional system (again, like any

grammatical system) is susceptible to a new kind of contact-based semantic shift.

Thus Mr. Pete’s refusal of a fictional landscape created an opportunity to study the

51

use of Ahtna directionals by a representative of the bilingual portion of the Ahtna

community in situ, in reference to actual locations that are familiar to him.

3.3 Forming the Hypothesis

The first indication that the Ahtna directional system may be undergoing

semantic change based on contact with English occurred midway through the first

elicitation session. The stimulus at the time was the diorama as pictured in Figure 3.1

above. As we have just seen, the river in the diorama had been established as the

Tazlina River, which flows in an easterly direction to its mouth at the Copper River

(Map 3.1, 5F-G). In the diorama this is toward the bottom of Figure 3.1 (the Copper

River is not represented). Thus, in terms of the cardinal directions represented in the

diorama, Mr. Pete is sitting to the north of the Tazlina River, and I am sitting to the

south.

It is also important to understand the actual cardinal directions of the room in

which the sessions take place, because at times both Mr. Pete and I refer to these

directions for clarification. In this and in all other examples from these four

fieldwork sessions, the cardinal directions of the room, show in Figure 3.1 below, are

as follows: the video camera had been placed in the southern edge of the room and

points north, thus east is to the reader’s right and west is to the reader’s left. In all

examples in this chapter and the next, however, descriptions of gestures indicating

direction refer to the directions understood by the discourse participants (Mr.

Pete and myself) to be represented in the diorama unless otherwise specified.

These were established when the each diorama was being arranged on the table, and

52

reflects our shared knowledge of the regions depicted. The directions of the diorama

are are in Figure 3.1, Figure 3.4, Figure 3.5, and Figure 3.6.

Figure 3.2: Cardinal Directions of the Room

in Which the Fieldwork Took Place

Approximately fifteen minutes into the session, Mr. Pete used the English words

north and south for the first time, after having used exclusively Ahtna terms (nor had

I used those words that day either). At the time I was testing the integrity of Major

River Orientation Principle 2. In (3.2), I had just placed the figurine of a hunter with

a rifle to the south of and facing away from the river, with a toy moose in its line of

sight. While this direction is upland of the Tazlina, if Principle 2 is intact in Mr.

Pete’s use of Ahtna directionals, we would expect him to call this daa’ ‘downriver’,

in relation to the flow of the larger Copper River.3

3 For this and all directions in this chapter and in Chapter 4, the stem morpheme

of the directional is the only morpheme that is relevant to the discussion. Readers may ignore affixes for now and rely on the English gloss to decode the stem that is

53

(3.2) Use of English north and south in elicitation of Ahtna directionals; diorama: the Tazlina River

01 AB; Would you call this, (1.4)

02 um, (0.3)

03 ’ungge? DIST.upland.ALL

‘Would you call this ‘to distantly upland’?’

(0.6) 04 MP; ’Ungge ((sic)) yi’e te dic’,

DIST.upland.ALL 3S.with 3S.SUB.shoot.PERF

05 ehn-[heh]. ‘He shot it toward the distant upland, yes.’

06 AB; [’Ungge], DIST.upland.ALL

‘To distantly upland,’

07 Okay. (0.5)

08 Alright. 09 MP; It’s the south,

(0.2) 10 ’udaa’a.

DIST.downriver.ALL

‘To distantly downriver.’

(0.4) 11 AB; ’Udaa’a.

DIST.downriver.ALL


(0.3) 12 MP; Ehn-heh.

(0.2) 13 AB; What if he’s like this.

(0.2) 14 MP; North.

(0.6) 15 AB; Okay how do you say that. 16 MP; ’Un’e.

DIST.upriver.ALL

‘To distantly upriver.’

present in each directional. The affixes are the topic of discussion in Chapters 6 and 7.

{AB gestures upland/south}

{MP points to south}

{AB orients toy moose

to downland/north of toy hunter}

54

(0.2) 17 AB; ’Un’e.

DIST.upriver.ALL


18 MP; ’Unuuxe c’e-, DIST.upriver.AREA ABAN

(0.6) 19 C’ena’ ts’idi ’aa de xu

INDEF.POSS.river keep.linear.O.in.position SP/TEMP.ADVZ AREA

su ize gh[aen le’]. EMPH 3S.SUB3S.OBJ.kill.sg.obj.PERF indeed

‘In an area distantly upriver, where the river extends out, he kills it (the moose) right there.’

20 AB; [Uh-huh]. (1.2)

21 MP; Xu ize ghaen means, AREA 3S.SUB3S.OBJ.kill.sg.obj.PERF

22 they kill moose there. 23 AB; Xu ize ghaen.

AREA 3S.SUB3S.OBJ.kill.sg.obj.PERF

‘He kills it there.’

24 MP; Mm-[hm]. 25 AB; [Okay].

(1.2) 26 AB; And did you just call this

4 north?

(1.1) 27 MP; I would call north [here

5],

28 AB; [Mm-hm]. (2.4)

29 MP; that’s north, 30 right here

6.


In IU 01, I erroneously suggest the term ’ungge ‘to distantly upland’ to indicate

the direction away from the shore of the Tazlina River, which Mr. Pete at first

accepts in IU 04. In IUs 07 and 08, I acknowledge that I understand and accept the

4 The referent of this in IU 26 is the north direction in the diorama.

5 The referent of here in IU 27 is the north direction in the diorama.

6 The referent of here in IU 30 is north in the room.

{AB gestures north across river}

{MP gestures north on diorama, then north in room}

55

referenced direction as ’ungge ‘to distantly upland’. However, Mr. Pete immediately

recognizes our agreed-upon error in IUs 09 and 10, and corrects himself and me,

saying that the direction is actually south, which he then translates as ’udaa’a ‘to

distantly downriver’. Referring to this direction as ’udaa’a in Ahtna is to be

expected in accordance with Major River Orientation Principle 2: the Tazlina River

is a tributary of the Copper River, which is the major river in the watershed and

flows, in this region at least, from north to south (Map 3.1, 5G-J). What is

interesting, however, is that Mr. Pete uses this term as the translation of the cardinal

English south (IU 09), rather than referring in any way to the flow of the nearby

Copper River (i.e., he did not say in English downriver in IU 09).

In IU 13 I reverse the orientation of the figurines so that the moose is downland

and to the north of the hunter. Mr. Pete immediately again uses an English cardinal

term, north, to describe, and also translates as ’un’e ‘to distantly upriver’. Again, the

use of ’un’e ‘to distantly upriver’ is in accordance with Major River Orientation

Principle 2, but the fact that Mr. Pete immediately provides a translation as English

north is indicative of the bilingual teacher-student interaction. When I ask for

clarification in IU 26 (did you just call this [i.e., north on the diorama] north?), Mr.

Pete confirms that he indeed intended to say north by gesturing toward that direction

both in the diorama and in the room. There is little doubt that Mr. Pete (i) is not

confused by the directions in the diorama in relation to the actual directions in the

room, and (ii) equates, at least to some degree, the notion of ‘north’ with nae’

‘upriver’.

56

Given that Mr. Pete is a first language speaker of Ahtna, a reasonable null

hypothesis – and the one I entered the research program with – would be that he

keeps the riverine directional system of Ahtna separate from the cardinal system of

English. However, the evidence presented in (3.2) lead me to form the alternate

hypothesis that for Mr. Pete the two systems are not completely distinct from one

another, and that their contact with one another in my consultant’s daily experience,

as well as his daily interaction with non-native speakers of Ahtna (not only myself,

but most members of the community) is causing a change of meaning in the Ahtna

system. In other words,

H0: There is no contact-based semantic change in the Ahtna directional system; the directional systems of English and Ahtna are separate and speakers do not conflate the two to any degree.

H1: There is contact-based semantic change in the Ahtna directional system: at least in bilingual settings, speakers conflate the directional systems of English and Ahtna in their speech to some degree (i.e, when speaking with a non-native speaker of Ahtna).

Disproving the null hypothesis in a bilingual setting would provide evidence for

a contact-induced semantic shift in which the different bases of the frames of

reference for the two systems become more similar to some degree.

3.4 Evidence for Change: nae upriver Becoming Equated with

north

While Mr. Pete’s translation of south as ’udaa’a ‘to distantly downriver’ and

north as ’un’e ‘to distantly upriver’ allows for the formation of H1, it is not sufficient

57

to allow us to reject H0, since it does not tell us how Mr. Pete assigns directional

terms across larger regions. In order reject H0, we need some evidence that the

principles of Major River Orientation, the hallmark difference between the two

systems, are not fully intact. In other words, we need evidence that Mr. Pete does not

always (e.g. when interacting with non-native speakers of Ahtna) shift the major axis

of the spatial grid when transitioning from one drainage to another (Principle 1),

and/or that he does not always assign to minor rivers the spatial grid of the major

river (Principle 2).

I tested Major River Orientation Principle 1 in the second elicitation session by

using two dioramas, one representing the Matanuska River drainage (Map 3.1, 2-

3G), and another representing the adjacent Copper River Drainage (Map 3.1,

between E5-7 and J5-7). These locations are ideal for testing Principle 1 because the

Matanuska River runs in a drainage located directly west of the Copper River

drainage, and flows roughly perpendicular to the Copper. It begins at a glacier in the

east and flows to the west and southwest into the salt water of the Knik Arm of Cook

Inlet. The Glenn Highway, the only road connecting the Copper Valley to

Anchorage, runs parallel to the Matanuska River along its north bank and is very

heavily traveled by residents of the Copper River valley, including Mr. Pete.

If the Matanuska River drainage is perceived by speakers to be a separate

drainage from that of the Copper River, we would expect that in accordance with

Major River Orientation Principle 1, the main axis of the spatial grid would shift to

match the flow of the Matanuska, with nae’ ‘upriver’ now corresponding to east.

Indeed, the fact that the Matanuska drainage is a distinct drainage is supported by

58

two arguments. The first is a physical argument: the change in terrain between the

two drainages is obvious to anyone traveling along the Glenn Highway. Traveling

westward toward Tahneta Pass, the elevation gain is gradual, but once one is over

the pass, the terrain drops off dramatically and one finds oneself descending quickly

though the deep river canyon. Figure 3.3 shows the elevation along the Glenn

Highway; Tahneta Pass is at approximately mile 67.

Figure 3.3: Elevation Profile of the Glenn Highway

The second argument for the clear distinction between the Copper and

Matanuska drainages is based in Ahtna language and culture: the last point at the top

of the pass from which the Copper River can be seen is named Nekets’alyaexde

‘where we turn around’. At this location, travelers would traditionally stop “to look

back and say a prayer” (Fred Ewan, p.c..; Map 3.1, G-4). The separation between the

two drainages is part of shared Ahtna culture, and thus it is expected that Major

River Orientation Principle 1 would cause a shift of the main axis of the spatial grid

between the two regions.

59

The diorama representing the Matanuska River drainage is shown in Figure

3.4. The directions embodied in the diorama are as follows: east is toward the bottom

of Figure 3.4, west is toward the top, Mr. Pete is sitting to the north and I am to the

south. The blue fabric represents the Matanuska River, and the black fabric

represents the Glenn Highway.

Figure 3.4: Diorama of the Matanuska River Drainage

The expectation of Major River Orientation Principle 1 is that Mr. Pete will refer

to the upriver/east direction with some form of nae’ ‘upriver’, because the

Matanuska originates in the east and is a separate drainage from the Copper. Just

prior to the excerpt in (3.3) below, Mr. Pete had instead referred to this direction

with ngge’ ‘upland’ (which I discuss in section 0 below); in IUs 01-04 I confirm that

the east direction here is not a form of nae’ ‘upriver’.

(3.3) Nae’ ‘upriver’ only used for north-south rivers; diorama: the Matanuska River 01 AB; This isn’t, 02 ’uniit?,

DIST.upriver.PUNC

03 AB; or ’un’e, DIST.upriver.ALL

{AB indicates toy dog running upriver/east}

60

04 AB; ’unae’?. DIST.upriver

‘This isn’t ‘at a place distantly upriver’ or ‘to distantly upriver’ or ‘distantly upriver’?’

05 MP; ## 06 Some-,

(0.6) 07 except, 08 it’s e:hn,

HES

(0.3) 09 north they’d say ’uniit.

DIST.upriver.PUNC

‘[If] it’s (i.e., if the river were flowing from the) north, they’d say ‘a point distantly upriver’.’

(0.2) 10 AB; O:h okay.

(0.8) 11 Only if this is going north. 12 MP; Ye[ah, 13 AB; [Okay], 14 MP; that’s

7] going,

(0.7) 15 west ((sic)), 16 AB; we’ll do that one in a minute, 17 MP; going so-- 18 east. 19 AB; Okay.


Mr. Pete responds that ’uniit ‘to distantly upriver’ would be an appropriate

response only if the river were flowing from the north to the south (IUs 06-09). His

first turn consists of the abandoned some- in IU 06 (which I interpret as being a

truncation of sometimes), a hesitation in IU 08 while he makes a gesture indicating a

north-south river that is perpendicular to the Matanuska, and the statement in IU 09

that if the river were coming from the north, one could say ’uniit ‘to distantly

{MP gestures as if the river were flowing north-south}

{AB indicates direction of flow of river}

61

upriver’ for the upriver direction. ’Uniit is presumably not correct here for Mr. Pete

because the Matanuska flows east to west (IUs 14-18). I confirm this in IU 11, in

which I am gesturing a river flowing from north to south (which is the referent of

this in IU 11). Mr. Pete’s gesture and statement that ’uniit ‘to distantly upriver’

would be appropriate if the upriver direction were to the north shows that Mr. Pete is

in fact aware of the connection between river direction and the Ahtna directional

terms. However, for him ‘upriver’ can only be used when the river flows from north

to south.

The reason for Mr. Pete’s association between nae’ ‘upriver’ and north is likely

due to the prominence of the Copper River in Mr. Pete’s home region. For residents

of the Central Ahtna dialect area, which includes the villages of Gakona, Gulkana,

Tazlina, and Copper Center, the Copper River flows from north to south. Thus the

opportunities to refer to the upriver direction as north (and vice versa) in English

conversation are abundant, and in the next example Mr. Pete goes on to show that his

association between the two is quite strong.

Example (3.4) is based on work with the diorama of the Copper River drainage.

Notice that while the Copper River does flow southward for most of its 287 miles, at

its headwaters it flows nearly due north out of the Copper Glacier before quickly

turning counterclockwise around the base of Mount Sanford. The diorama shown in

Figure 3.5 corresponds to the length of the entire Copper River (the blue fabric), the

Richardson and Tok Highways (the black fabric), and the Wrangell Mountains (the

7 The referent of that in IU 14 is the Matanuska river.

62

green mountain figurine). Our shared knowledge is that the town of Tok (Map 3.1,

D7), not depicted, would be off the top right corner of the table.

Figure 3.5: Diorama of the Copper River,

With the Richardson and Tok Highways

In (3.4) below, Mr. Pete and I are first discussing highway travel along the

Richardson Highway, which runs parallel to the Copper River, in the area south of

the junction of the Tok Highway. The directionals Mr. Pete uses here are to be

expected: south, which is the downriver direction of the Copper, is ’udaa’a ‘to

distantly downriver’, and north, which is the upriver direction, is ’un’e ‘to distantly

upriver’. In IU 12 I ask Mr. Pete what to call the direction one takes when one turns

to the northeast on the Tok Highway. This is followed in IU 36 by a discussion of

traveling all the way to Fairbanks, some 250 miles north of the Copper River area.

63

(3.4) Strength of association between nae’ ‘upriver’ and north on the Tok and Richardson Highways; diorama: the Copper River

01 AB; Okay, (1.2)

02 how about, 03 the car,

(0.6) 04 driving on the Ri[chards-]- 05 MP; [’U]daa’a.

DIST.downriver.ALL

06 AB; on the Richardson, 07 MP; ’Udaa’a.

DIST.downriver.ALL

‘To distantly downriver, to distantly downriver.’

08 AB; ’Udaa’a. DIST.downriver.ALL


(2.2) 09 MP; ’Un’e.

DIST.upriver.ALL


10 AB; ’Un’e, DIST.upriver.ALL


11 okay. 12 and then you get to here,

(1.5) 13 what’s that direction.

(0.2) 14 On Tok ((i.e., Tok Highway).

(0.8) 15 MP; Ehn, 16 ’un-,

ABAN

17 ehn, 18 Tok ts’e’ kughiyaa .

T. to 3s.sub.sg.go.IMPERF.AREA

‘Um, he is going to Tok.’

(0.7) 19 AB; Tok ts’e’.

T. to

‘To Tok.’

{AB indicates toy car driving downriver/south on the Richardson Highway}

{AB turns car around so that it is going upriver/north toward Fairbanks}

{AB indicates the car is at the fork of the Richardson and Tok Highways, and heads upriver/northeast toward Tok}

{AB points at Tok}

64

(0.4) 20 MP; Ehn-heh.

(0.5) 21 AB; And what direction is this.

(0.7) 22 MP; I don’t know.

(0.5) 23 AB; Okay.

(0.7) 24 Okay. 25 MP; I don’t know.

(0.4) 26 AB; Okay, 27 you just say Tok. 28 MP; <SMILE>I can’t know. 29 AB; Okay, 30 that’s alright. 31 MP; No I don't know. 32 AB; <SMILE>Oh hush you</SMILE>. 33 MP; @@@@@@@</SMILE>. 34 AB; But this way is ’un’e.

DIST.upriver.ALL


(0.7) 35 MP; That’s uh,

(0.2) 36 MP; going to Fairbanks, 37 [’un]’e.

DIST.upriver.ALL


38 AB; [Yeah], 39 ’un’e.

DIST.upriver.ALL


(0.3) 40 So this one’s not ’un’e.

DIST.upriver.ALL

‘So this one’s not ‘to distantly upriver’?’

{AB gestures upriver/ northeast}

{AB moves car to Richardson Highway, then moves it upriver/north toward Fairbanks}

{AB moves car back to Tok Highway and indicates it moving upriver/northeast}

65

(1.4) 41 MP; I don't know what they, 42 AB; Okay. 43 MP; what you call that #, 44 you go west ((sic)). 45 AB; Okay.

(0.4) 46 MP; Eastwest ((sic)) I [guess]. 47 AB; [Yeah] north-- 48 northeast.

(0.7) 49 MP; Northeast. 50 AB; Northeast. 51 Okay.

(0.4) 52 MP; Northeast, 53 [that’s right]. 54 AB; [Okay]. 55 MP; That's the name.


On two fronts this rich example shows both the strength of the association

between nae’ ‘upriver’ with north, and the loss of integrity of Major River

Orientation. First, consider Mr. Pete’s use of upriver to refer to the route to

Fairbanks (Map 3.1, A3; IUs 34-39). Fairbanks is quite some distance from the

Copper River, across a major mountain range and nowhere near any southerly

flowing river. That Mr. Pete agrees with my use of this term (i.e., my use of it in IU

34, which was a reiteration of what he had told me in IU 09) to refer to Fairbanks is

striking evidence for the erosion of Major River Orientation Principles 1 and 2,

because the association with north effectively extends beyond the Copper River

region to refer to locations hundreds of miles away and with no association with an

upriver direction (indeed, even the upriver direction of the Tanana River is to the

66

east/southeast of Fairbanks).8 The English cardinal system allows for this kind of

extension – north is north regardless of intervening physiographic features – but in

Ahtna the term Ba’aaxe ‘a general area outside’ has been lexicalized to refer

specifically to the Tanana River region, including Fairbanks. Mr. Pete does not use

this term.

Second, and perhaps even more crucially, consider IUs 12-33, in which I ask Mr.

Pete about the direction to Tok. The Tok Highway follows the curve of the Copper

River around the north face of Mt. Sanford, and in this region the upriver direction is

to the northeast. After some hesitation (and the abandonment of a directional in IU

16), Mr. Pete refers to this direction as Tok ts’e’ ‘to Tok’ (IU 18). This is the first

time he uses a postpositional phrase instead of a directional, and upon further inquiry

(IUs 21-33 and 40-55) he comments that he simply does not know what to call it.

This suggests that Major River Orientation Principle 2 is losing its centrality to the

semantics of the Ahtna directional system: when the upriver direction on the Copper

is no longer north, Mr. Pete hesitates about which directional term to select and

instead opts for a postposition (I return to his solution to the problem of directional

reference near the headwaters of the Copper in section 3.5.3 below).

8 In example (3.4) Mr. Pete agrees to my suggestion of ’un’e as indicating the

route to Fairbanks, but my suggestion in line 34 was based on previous conversations in which he used the term in this way. I am referencing information he already gave me but that is not shown here.

67

3.5 The Role of General Topography

The evidence above supports the hypothesis that there is some degree of contact-

induced semantic shift occurring in Mr. Pete’s use of the Ahtna directional system, at

least when interacting with a second language learner of Ahtna. The fact that Mr.

Pete is unsure of what to say in cases in which upriver and north do not align

suggests that this is the beginning of a shift, rather than a mature shift, when we

would expect speakers to have fully integrated the new system into their grammar.

While Ahtna is so endangered it is likely we will never witness a complete shift,

it is still worth examining some of Mr. Pete’s strategies for dealing with directional

reference in cases that are now problematic because of the gaps left in the shifting

paradigm. Two of these are examined below.

3.5.1 Ngge upland in the Matanuska River Drainage

Let us return to the discussion of the axis of the directional grid in the Matanuska

River drainage (Figure 3.4 above). Recall that this river flows from east to west, but

that in (3.3), Mr. Pete rejected nae’ ‘upriver’ for east. Instead, as can be seen in (3.5)

below, he uses ngge’ ‘upland’ to refer to the east.

(3.5) Use of ngge’ ‘upland’ for the upriver direction; diorama: The Matanuska River drainage

01 AB; What direction is this? (0.8)

02 MP; ’Ungge nac’e’aa . DIST.upland.ALL 3S.SUB.INDEF.OBJ.handle.cpt.obj.ITER.IMPERF.PERAMB

‘S/he is driving back to distantly upland.’

(0.6) 03 AB; ’Ungge.

DIST.upland.ALL

‘To distantly upland.’

{AB moves car upriver/east}

68

04 MP; Ehn-heh, ‘Yes.’

05 AB; Okay, (2.0)

06 ’ungge, DIST.upland.ALL

‘To distantly upland.’

07 so, 08 if this river, 09 is frozen over?,

(0.7) 10 and he’s on top running?,

(0.4) 11 MP; Mm-hm, 12 AB; what direction is that. 13 MP; ten k’e.

ice on

(0.4) 14 AB; Mm-hm?. 15 MP; Ten k’e ngge’ nadzitez’aan.

ice on upland 3S.SUB.animal.runs.ITER.PERF

‘On the ice, it runs back upland on the ice.’

(1.5) 16 MP; Ten k’e ngge’ nadzitez’aan.

ice on upland 3S.SUB.animal.runs.ITER.PERF

‘It runs back upland on the ice.’


Mr. Pete uses ’ungge ‘to distantly upland’ to refer to a car driving in the upriver

direction on the highway in IU 06, and ngge’ ‘upland’ to refer to a dog running

upriver atop the frozen surface of the river in IUs 15 and 16. Before considering why

Mr. Pete selected these terms, let us look at his other strategy for avoiding the now-

inappropriate use of nae’ ‘upriver’ (i.e., inappropriate because nae’ is now reserved

for north-south rivers).

{AB moves toy dog upriver/ east atop surface of Matanuska River}

69

3.5.2 Tgge up (vertically) Along the Tazlina River to Tazlina Lake

The excerpt in (3.6) is based on a diorama of the trail system around Tazlina

Lake (Map 3.1, G4), including the Glenn Highway and the Lake Louise region (Map

3.1, F4). This diorama, which Mr. Pete had in fact assembled from memory, is

shown in Figure 3.6. The brown strip of fabric on the left represents an old trail that

used to run parallel to the Tazlina River on its south bank up to Tazlina Lake; the

blue fabric to the right represents the Tazlina River, which flows eastward from

Tazlina Lake, also pictured. The black fabric represents the Glenn Highway. Again,

the bottom of the figure represents east; Mr. Pete is sitting to the diorama’s north,

and I am to the south.

Figure 3.6: Diorama of the Trail to Tazlina Lake

In (3.6), we are discussing foot travel along the trail toward Tazlina Lake. Mr.

Pete refers to this direction as ’utgge ‘to distantly up (vertically).’ This is notable

because the terrain in this area is not overwhelmingly vertical.

70

(3.6) Tgge’ ‘up (vertically)’ along Tazlina River; diorama: the trail to Tazlina Lake 01 AB; So if he’s walking on the trail,

(0.8) 02 how would you say, 03 he’s walking that way.

(0.8) 04 MP; ’Utgge natesdyaa.

DIST.up.ALL 3S.SUB.sg.go.IMPERF.ITER.INCEP

‘S/he is starting to go back/again to distantly up (vertically).’

05 AB; ’Utgge, DIST.up.ALL

‘To distantly up (vertically).’

(0.7) 06 cause he’s going up to the, 07 MP; Yeah. 08 AB; lake? 09 [Okay], 10 MP; [Going] up to the lake.

(1.2) 11 AB; Okay.

(0.2) 12 MP; ’Utgge ts’ina’idyaa.

DIST.up.ALL 3S.SUB.sg.go.IMPERF.straight.ITER

‘S/he is going straight back to distantly up (vertically).’

13 AB; Ts’ina’idyaa. 3S.SUB.sg.go.IMPERF.straight.ITER

‘S/he is going straight back.’

14 MP; Mm-hm. ((Markle Pete, oai.paradisec.org.au:ALB01-060, 00:10:00.835-00:10:17.615))

Mr. Pete answers in IU 04 that the correct direction here is ’utgge ‘to distantly up

vertically, which I echo in IU 05, and then ask for a reason in IUs 06 and 08. He

confirms that the reason for ’utgge ‘to distantly up (vertically)’ is that he’s going

‘up’ to the lake (i.e., the English polysemous up that includes nonvertical directions,

see Section 5.3). A few seconds later, Mr. Pete confirms that this direction is ’utgge

‘to distantly up (vertically)’, shown in (3.7), in response to my verification in IU 03

{AB moves toy hunter upriver/west along a trail parallel to Tazlina River toward Tazlina Lake}

71

that ’un’e ‘to distantly upriver’ is not possible, even though this is the upriver

direction on the Tazlina River.

(3.7) Confirming tgge’ ‘up (vertically)’; diorama: the trail to Tazlina Lake 01 AB; Did you also say ’un’e?

DIST.upriver.ALL

‘Did you also say ‘to distantly upriver’?’

(1.1) 02 MP; Huh?

(0.4) 03 AB; Can this be [’un’e]?

DIST.upriver.ALL

‘Can this be ‘to distantly upriver’?’

04 MP; [Hehn-eh], ‘No.’

05 AB; No?, (0.3)

06 MP; ’utgge. DIST.up.ALL


07 AB; ’Utgge, DIST.up.ALL


08 okay. ((Markle Pete, oai.paradisec.org.au:ALB01-060, 00:10:30.783-00:10:38.223))

3.5.3 Why ngge and tgge ?

Why does Mr. Pete select ngge’ ‘upland’ and tgge’ ‘up (vertically)’ as alternate

strategies to nae’ ‘upriver’ for referencing direction? First, let us consider the notion

of a perpendicular spatial grid as part of a directional system. We might be tempted

to force the Ahtna riverine directional system into a perpendicular grid like the

English cardinal system: just as the north-south axis is 90 degrees to the east-west

axis, it seems natural to conceive of the nae’ ‘upriver’-daa’ ‘downriver’ axis as

{AB points upriver/west along Tazlina River trail}

{AB moves toy hunter in same direction}

72

being 90 degrees to the ngge’ ‘upland’-tsen ‘downland’ axis. A prototypical river

flows through, and indeed, creates, a valley; a person walking perpendicularly away

from such a river would of course be walking up a slope.

But there is no reason to think that such slopes only exist perpendicularly to a

river, and experience with actual terrain tells us they usually do not. Furthermore, for

semi-nomadic cultures, reference to elevation that needs to be surmounted (or

circumambulated) by foot or dogsled may be more important than strict adherence to

a perpendicular grid, which is most useful for marine navigation. In this sense, the

Ahtna directional system is more three-dimensional than the English system,

including changes in elevation in the same paradigm as axial relations.

As for Mr. Pete’s selection of ngge’ ‘upland’ to refer to the east/upriver direction

on the Matanuska, recall that this is an area of steep elevation change. The salience

of the sloping topography makes ngge’ a reasonable choice when nae’ ‘upriver’ is

not available, as it now is not available for the westerly-flowing Matanuska because

he now associates this term with north. While choosing ngge’ ‘upland’ to refer to an

upwardly sloping direction is not necessarily innovative, reserving nae’ ‘upriver’

only for rivers that flow from north to south is innovative, and a sign of a semantic

shift.9

Mr. Pete’s use of tgge’ ‘up (vertically)’ in (3.6) and (3.7) to describe an area that

is not particularly vertical is intriguing. Although tgge’ and its opposite igge’ ‘down

9 Kari notes the fallacy of assuming that ngge’ ‘upland’ refers only to directions

that are 90 degrees from the river is apparent in the lexicalizations of directionals in

73

(vertically)’ are morphologically members of the lexical class of directionals, they

do not have an exclusively topographic semantic basis. These terms can refer to

many kinds of steep verticals including ladders, cliffs, and trees. In (3.8) Mr. Pete

uses tgge’ ‘up (vertically)’ to refer to the vertical cliff in Frog, Where Are You?

(Meyer 1969), and in (3.8) he uses it to describe a squirrel’s perch at the top of a

tree.

(3.8) Tgge’ ‘up (vertically)’ for true verticals

(a) MP; ’Utgga hwnene u’e

DIST.up.PUNC hill 3S.OBJ.with

niyitats’ 3S.SUB.animal.rushes.PERF.TERMIN

ic’ae u’e dadits’et tuu yii dog 3S.OBJ.with 3S.SUB.animate.moves.indep.down.PERF water in

tahghidaek. 3PL.SUB.pl.obj.water.move.indep.PERF

‘Up on a cliff it stops suddenly; he falls with the dog and they fall in the water.’

((Markle Pete, Naghaay Ndaane Zidaa ‘Frog Where Are You’, oai.paradisec.org.au:ALB01-042, p.21-22))

(b) 01 MP; Dligi tgge’ dazdaa.

tree.squirrel up 3S.SUB.sg.sit/stay.up.IMPERF

‘The tree squirrel is sitting high up (vertically) on something.’

((…)) 02 ’Utggat adazdaa.

DIST.up.PUNC 3S.SUB.sg.sit/stay.up.IMPERF

‘It is sitting at a point distantly high up (vertically) on something.’


toponyms: “A river name + ngge’ means the entire drainage, and these can be an array of 360 degrees, smaller streams as well as larger streams” (James Kari, p.c.).

74

In both of these examples, Mr. Pete uses directionals based on the stem tgge’ (the

additional morphology is not relevant to the present discussion). Using tgge’ to

describe directions on the ground as well as for vertical height can also be seen as an

example of the Ahtna directional system incorporating the third dimension, but this

time with perhaps some influence from English. English up encompasses a broad

and very general semantic range, e.g. upriver, uphill, uptown, up the road, etc. Given

that Mr. Pete is, like all other living speakers of Ahtna, bilingual in English, it is

reasonable to suggest that perhaps tgge’ ‘up (vertically)’ is now being used to

describe situations in which English up would be appropriate, rather than just to refer

to strictly vertical.10

Incidentally, tgge’ ‘up (vertically)’ is how Mr. Pete tentatively solves the

problem of referring to the upriver direction on the curved upper portion of the

Copper River. (3.9) begins with a discussion of a dog running up the frozen river in

the north-south portion of the Copper. In IU 16 Mr. Pete uses ’utgge ‘to distantly up

(vertically)’ to describe the dog’s change of direction when he heads first northeast,

then east and finally south to reach the headwaters. Although I confirm the transition

from nae’ ‘upriver’ to tgge’ ‘up (vertically)’ twice in this excerpt (IUs 22-31 and 32-

34), in the end Mr. Pete is still not completely satisfied and comments that he simply

does not really know what to say (IU 37).

10 Ahtna also has several far more grammaticalized – and hence semantically

general – adverbial verb prefixes corresponding to English up, e.g., da- (dazyaa ‘s/he went up’) and ka- (kaghi’aa ‘linear object extends up’). Tgge’ is still phonologically independent and behaves as an affix-taking member of the directional lexical class,

75

(3.9) Tgge’ ‘up (vertically)’ on the upper Copper River; diorama: the Copper River 01 AB; One more question about ic’ae,

dog

(0.7) 02 when he’s running on top of the frozen river,

(0.3) 03 this [was, 04 MP; [’Un’e],

DIST.upriver.ALL


05 AB; ’un]’e, DIST.upriver.ALL


06 MP; ’un’e dza’aa . DIST.upriver.ALL 3S.SUB.animal.runs.IMPERF

‘It is running to distantly upriver.’

07 AB; Yeah. (0.3)

08 Is this also ’un’e dza’aa ?. DIST.upriver.ALL 3S.SUB.animal.runs.IMPERF

‘Is this also ‘it is running to distantly upriver’?’

(0.3) 09 MP; Huh? 10 AB; Is this the same thing?, 11 ’un[’e]?

DIST.upriver.ALL

‘To distantly upriver?’

12 MP; [Ehn]-heh. ‘Yes.’

13 ’Un’e ta kuts’ighi’aan. DIST.upriver.ALL among clf.cpt.obj.IMPERF.AREA

‘It is (going) to distantly upriver.’

(0.2) 14 AB; What happens when he comes [here], 15 MP; [’Ut]gga ben ts’e’ sut’aen de.

DIST.up.PUNC lake to be.thus SP/TMP.ADVZ

‘A point distantly up (vertically) to where the lake is.’

which further suggests that its semantic extension to describe landforms that are not particularly vertical is a new development.

{AB moves toy dog upriver/ north}

{AB moves dog upriver/northeast around curve in Copper River}

{AB moves

dog upriver/east> south around curve in Copper R.}

76

(1.4) 16 AB; ’Utgge ben ts’e’,

DIST.up.ALL lake to

‘Distantly up (vertically) to the lake.’

17 MP; Ehn-[heh, ‘Yes.’

18 AB; [up to the, 19 MP; lake is here you know], 20 AB; up to the lake]. 21 MP; Copper Lake.

(0.8) 22 AB; So this is ’un-,

ABAN

(0.2) 23 ’un’e,

DIST.upriver.ALL

(0.2) 24 ’un’e,

DIST.upriver.ALL

25 right, ‘So this is ‘to distantly upriver’, right?’

(0.5) 26 MP; Yeah.

(0.2) 27 AB; is it still ’un[’e when],

DIST.upriver.ALL

‘Is it still ‘to distantly upriver’ when…’

28 MP; [’Ut]gge. DIST.up.ALL


(0.4) 29 AB; so now it’s ’ut-,

ABAN

30 MP; ’Utgge. DIST.up.ALL


31 AB; ’Utgge. DIST.up.ALL


(1.1) 32 AB; So this, 33 is not ’un’e.

DIST.upriver.ALL

‘So this is not ‘to distantly upriver’?’

{MP points to lake at head of Copper River}

{AB moves dog upriver/northeast around curve in Copper River}

{AB moves dog upriver/east>south around curve in Copper R.}

{AB indicates portion of Copper River where upriver is east and south}

77

(0.2) 34 MP; Ehn-heh.

‘Right.’

(0.4) 35 AB; You [couldn’t s]-, 36 MP; [I] don’t know. 37 AB; You don’t know, 38 okay.

(0.3) 39 Okay.


The semantic shift that is occurring in Mr. Pete’s use of the directionals is

summarized in Figure 3.7.

Figure 3.7: Diagram of Semantic Shift

As nae’ ‘upriver’ becomes strongly associated with English ‘north’ due to the

prevalence of the north-south flowing Copper River, it is no longer available to Mr.

Pete for ‘upriver’ in rivers that run in a different direction. This leaves a gap in the

paradigm, which Mr. Pete fills with either ngge’ ‘upland’, possibly based on the

slope of the land; or with tgge’ ‘up (vertically)’, possibly based on the semantic

range of ‘up’ in English, a language in which Mr. Pete is bilingual.

78

3.6 Conclusion

Although Ahtna is Mr. Pete’s first language, a lifetime of full bilingualism with

English is causing the erosion of Major River Orientation in his use of directionals.

Evidence for this erosion can be found in his reluctance to use nae’ ‘upriver’ for

rivers (or portions thereof) that do not flow north to south, and his extension of nae’

to refer to distant locations like Fairbanks.

Mr. Pete’s innovative use of Ahtna directionals indicates an ongoing shift of the

semantic basis of the system. The first change in this shift is the strong association of

nae’ ‘upriver’ to north. Although this step is not complete – we have seen that Mr.

Pete is still aware of the connection between nae’ and river flow – it nonetheless

leaves a semantic gap in the paradigm that other terms (ngge’, tgge’) must now be

recruited to fill. It is likely that Ahtna will become extinct before the entire system

shifts sufficiently that speakers no longer experience moments of hesitation about

which terms to select, but this chain-shift pattern is a common indicator of contact-

based language change (e.g. Thomason & Kaufman 1988, Heine & Kuteva 2005).

Important to this analysis is the understanding that during the fieldwork sessions

described here, Mr. Pete was working to accommodate a learner of Ahtna whose first

language is English. In fact this situation mirrors most of Mr. Pete’s opportunities to

use Ahtna in his family and community. There are few fluent speakers of Ahtna

today, and Mr. Pete spends most of his day communicating to learners or

nonspeakers of Ahtna. These daily contacts influence the way Mr. Pete adapts his

knowledge of the indigenous semantics of the Ahtna directional system to

79

accommodate English speakers who have a very different understanding of the

semantics of direction.

It is also crucial to stress that Mr. Pete’s shifting use of directionals does not

reflect in any way an ignorance of Ahtna language or the semantics of the directional

system; to the contrary, he is fluent in the language and very knowledgeable about

the geography of the Ahtna territory. Rather, it reflects the fact that a very real

contact-induced change is taking place in the language. In terms of Thomason &

Kaufman’s table of the linguistic results of language contact (1988:50), Mr. Pete’s

hesitation is typical of the early stages of intense contact leading to large-scale

language shift, in which influence between the substratum and superstratum is

moderate to heavy. The semantic change we observe in Mr. Pete’s use of the

directionals is a natural result of intense language contact, even though the contact

could arguably be viewed as unnaturally catalyzed by historical pro-English policies

and economic pressure (a discussion that is beyond the scope of the present work).

Regardless of the cause of the contact between the two languages, my interactions

with Mr. Pete in this study are akin to those taking place within the changing

demography of the Ahtna-speaking community.

Looking ahead, the next chapter discusses the refinement of the difficult

semantics of one directional stem, nse’, which has traditionally been translated as

‘ahead’.

80

Chapter 4

Refining the Semantics of nse’ ‘ahead’

In this chapter I discuss the semantics of one directional stem, nse’, which has

traditionally been glossed as ‘ahead’. In 4.1, I introduce the research problem,

presenting both Kari’s (2010) opinion on the semantic complexity of this stem, and

my own difficulties with assigning meaning to it in my fieldwork. In Section 4.2, I

present five possible explanations for the extension of nse’, three of which come

from my consultant’s intuition. In each case, the explanation fails to account fully for

the patterning of nse’ with some locations but not with others. I discuss a more

unified explanation in Section 4.3, in which a combination of features together

contribute to the complex semantics of this stem and also lend more evidence to the

idea that the Ahtna directional paradigm takes elevation into account. This is

different from the English directional paradigm, which only accounts for changes in

two dimensions.

4.1 Semantic Complexity of nse

In a discussion about the semantic basis of the directionals, Kari writes, “nse’ is

the most interesting and semantically complex directional” (2010:132).1 Leer (1989)

1 Kari (p.c.) also notes the temporal senses of this stem: “Not so prevalent in

Ahtna is the temporal meaning ‘ahead, in the future’, as in Gwich’in yeendoo ‘in the future’. [Ahtna] c’endze’, the term for ‘frail elder’ has this root, i.e., “from the future.” Quite interesting. I keep thinking about this root and it is truly amazing.”

81

reconstructs the stem as Pre-Proto-Athabascan *n sd > Proto-Athabascan *n s-, and

gives it two directional meanings, ‘ahead’ and ‘out into open water’. Synchronically,

the root upon which the stem is based can be used in constructions with a range of

meanings, including ‘forward, ahead, in front’ (’unse zidaa ‘(you) sit in front (e.g. of

car, boat)’), ‘over a fire’ (’unse na’sdulnaats ‘let’s prepare it over a fire’), ‘off from

shore, out into water’ (nse’ bey’tni ’aa ‘he is holding something (boat) off from

shore with a pole’), and ‘worse’ (konse dayaak ‘his health deteriorated’); when used

to refer to geography, it can mean ‘outside the Copper River valley’ (’unse i’e

dinaadyaa ‘he went back outside the Copper River valley with it’) or even ‘outside

Alaska’ (’unsghu denae uts’en c’ezdlaen ‘he got wealthy (lit. ‘he became a chief’s

son’) from it outside (of Alaska)’ (Kari 1990:314).

Within the directional system paradigm, Kari notes that the term participates in

three sets of orthogonal semantic oppositions (see Chapter 2) at different spatial

scales:

[…] nse’ occurs in three pairs of oppositions. Inside and outside […] nse’ patterns opposite ’ane’ [‘out, beyond, other side’]. The position of the hearth is the semantic thread for nse’::’ane’. The hearth was in the center of the house, and persons face the fire in both indoor and outdoor settings and at three distinct scales: (a) indoors: to the fire versus outdoors; (b) out in the local range: into the open, to the front versus over, beyond (a hill, mountain); and (c) long range (over the Chugach Mountains to the south::over the Alaska Range to the north) to the fire, to the front.

2

Rounding out the semantic oppositions for nse’ […] for intrinsic shapes (boats, any shaped [sic] object) and

2 In Chapter 3 we saw that for Mr. Pete, the direction over the Alaska Range is

not ’ane’ ‘out, beyond, other side’, but is now nae’ ‘upriver’.

82

anatomy nse’ is ‘front’ and n’e, nii is ‘back’, thus the dual meaning for n’e, nii is ‘upstream; back.’ In lakes and large streams the opposition is nse’::ngge’ ‘out on water; to the shore, to the upland’. (Kari 2010:132)

In other words, Kari is here defining the difficult semantics of nse’ in part based

on antonymy. One of its antonyms is ’ane’, and the antonymic relationship between

the two has slightly different meanings depending on whether it is applied indoors,

outside on a local scale, or outside on a larger, geographic scale. Another of its

antonyms is n’e/nii (= nae’), in which case the pair take on an antonymic

relationship of ‘front:back’ in animal anatomy and certain objects like rifles or boats.

Another antonym of nse’ is ngge’, in which the pair expresses an opposition between

the direction toward the open water versus the direction toward the shore.

As my field research on directionals progressed, it became clear to me that nse’

‘ahead’ at least applies to locations at or beyond the periphery of Ahtna territory.

Four easily-referenced established locations are potential candidates for being in the

nse’ direction: Fairbanks to the north (Map 3.1, A3), Anchorage to the west-

southwest (Map 3.1, H1), Valdez to the south along the Richardson Highway (Map

3.1 H4), and Cordova to the south near the mouth of the Copper River (not reachable

by road; Map 3.1, I5). During all of our work together, Mr. Pete never vacillated

about which towns were and were not in the nse’ ‘ahead’ direction. These are shown

in Table 4.1.

83

Table 4.1: Locations That Are and Are Not in the nse’ ‘ahead’ Direction

Fairbanks Anchorage Valdez Cordova

nse’ ‘ahead’? no:

nae’ ‘upriver’ yes yes

no: daa’

‘downriver’

It was not immediately clear to me what criteria determine which locations are

considered to be nse’ ‘ahead’, and upon further inquiry I found that the criteria were

not fully accessible to Mr. Pete either. Over the course of our work together, he

offered three possibilities for the semantic basis of nse’ ‘ahead’: that the term (i)

refers to locations that are further away than e.g. daa’ ‘downriver’, (ii) refers to

locations found on salt water, or (iii) refers to locations that are beyond a

mountainous divide, what Mr. Pete refers to as the “hump” or “ball” of the world.

These, as well as some possibilities from the literature, are discussed in turn below.

4.2 Possible Meanings

4.2.1 Meanings Arising From Fieldwork

4.2.1.1 Further Away Than un e

My fieldnotes (oai:paradisec.org.au:ALB01-AHTNB001-151.tiff) document a

discussion from late January 2010 that Mr. Pete and I had on the semantics of Ahtna

directionals, which ultimately led to the development of the videotaped research

program described in Chapter 3. This discussion involved the creation of a hand-

drawn map of the Ahtna region and some labeling of directions. Mr. Pete instructed

84

me to label Anchorage as ’unsghu ‘a general area distantly ahead’, with the

transcribed comment “way over” – this directional is based on the stem nse’.

Similarly, Fairbanks is labeled with ’unuuxe ‘a general area distantly upriver’, with

the transcribed comment “short distance” – this directional is based on a different

stem, nae’ ‘upriver’. See Figure 4.1, in which the relevant annotations are indicated

with arrows.

85

Figure 4.1: Fieldnotes Containing References to Relative Distances of nse’

and nae’ (oai:paradisec.org.au:ALB01-AHTNB002-151.tiff)

86

At that time, Mr. Pete expressed to me that his understanding was that nse’

‘ahead’ refers specifically to locations that are a great distance away (although how

‘great distance’ is defined is unclear), while other terms like nae’ ‘upriver’ were for

much shorter distances (Markle Pete, p.c.). It is crucial to note that this is not

actually true. Using the village of Copper Center as a starting point, Fairbanks is 259

driving-miles away, while Anchorage is closer, at 196 miles. Valdez is even closer at

105 miles. Cordova is not accessible by car, and requires either a 130-mile journey

down the Copper River (undertaken very rarely today, and usually only by sport

rafters) or a chartered flight – either way, certainly not ‘close’ in the sense of ‘easily

reachable’.3

Table 4.2 shows, despite Mr. Pete’s intuition that nse’ refers to far distances, how

these four locations compare in terms of their distance to the Copper River valley.

3 Interestingly, Mr. Pete’s intuition that nse’ refers to locations that are “at a far

distance” is reflected in a quote from Jules Jetté’s original lexical notebooks that is reproduced in the Koyukon dictionary: “With a direction (or location) sufficiently determined by the context or the circumstances, ne e [i.e., the Koyukon cognate of Ahtna nse’] simply means ‘further’, ‘further off’, and can apply to two opposite directions. Thus, one standing on the shore will use yoon e to designate the inland; and yoon ughe to designate the offing, in accordance with the thought which he presumes to be in the hearer’s mind” (Jetté & Jones 2000:467).

87

Table 4.2: Locations by nse’ and Distance


nse’ ‘ahead’ no


no daa’

‘downriver’

at a “far” distance

yes no no yes

4.2.1.2 Coastal Locations

Another option presented by Mr. Pete about the meaning of nse’ is that it refers

to locations found on a salt water coast. The association with the shore in this

interpretation reflects Leer’s reconstruction of the PA stem meaning ‘out into open

water’. Three of the four locations considered here are on the coast: Anchorage on

Cook Inlet, Valdez on Prince William Sound, and Cordova on the Gulf of Alaska

near the delta of the Copper River.

The lengthy excerpt in (4.1) comes from our work with the diorama of the

junction of the Richardson and Glenn Highways shown in Figure 4.2. The

Richardson Highway, which runs from Fairbanks in the north to Valdez in the south,

is depicted by the black fabric shown in the foreground of the figure, parallel to the

front edge of the table. The Glenn Highway, which runs from Anchorage in the west

to the Richardson Highway in the town of Glennallen in the east, is shown in the

figure perpendicular to the front edge of the table. In terms of the directions in the

diorama, Mr. Pete is sitting to the north and I am sitting to the south.

88

Figure 4.2: Diorama of the Junction of the Richardson and Glenn Highways

Example (4.1) is a discussion of the directions one uses to refer to Anchorage,

Fairbanks, and Valdez. (Cordova is not on the highway system and is therefore not

part of this conversation.) Early on Mr. Pete offers his intuition of the coastal

meaning of nse’ ‘ahead’ (IUs 08-10). He then contrasts this with Fairbanks, which is

’un’e ‘to distantly upriver’, calling Fairbanks “central” (IU 24).4 He then goes on to

reconfirm Anchorage and Valdez as ’unse ‘to distantly ahead’ because they are on

the natu’ ‘ocean’.

4 It is unclear in this example whether by “central” Mr. Pete means a central

distance (i.e., as opposed to a far distance, which is how I understood it at the time (IUs 27-35) based on our earlier conversation), or a central location in Alaska, i.e., not on the coast. Neither interpretation is more revealing than the other.

89

(4.1) Nse’ ‘ahead’ as ‘coastal’; diorama: the Richardson and Glenn Highways 01 AB; Anchorage, 02 is all the way down here,

(0.9) 03 right?,

(0.3) 04 MP; ’Unse.

DIST.ahead.ALL

‘To distantly ahead.’

05 AB; ’Unse. DIST.ahead.ALL


06 MP; Ehn-heh. 07 AB; What’s that mean.

(0.2) 08 MP; That means, 09 over,

(0.6) 10 towards the ocean.

(0.3) 11 AB; A:h, 12 towards the ocean, 13 MP; Ehn-[heh], 14 AB; [okay], 15 MP; wherever, 16 ocean,

(0.5) 17 AB; uh-huh,

(0.2) 18 MP; they say, 19 MP; ’unse ta.

DIST.ahead.ALL among


(0.3) 20 AB; ’Unse.

DIST.ahead.ALL


(0.6) 21 MP; ’Un’e in,

DIST.ahead.ALL

{AB indicates Anchorage far to the west}

{MP gestures to the real Anchorage outside the room}

90

22 Fairbanks, 23 that’s,

(0.7) 24 central.

‘ ‘To distantly upriver’ is Fairbanks, that’s central.’

(0.4) 25 AB; O[kay], 26 MP; [Cen]tral. 27 AB; oh right, 28 ’unse means,

DIST.ahead.ALL

(0.2) 29 further a[way, 30 righ]t?,

‘ ‘To distantly ahead’ means ‘further away’, right?’

31 MP; [Ehn-heh]. 32 AB; And ’un-,

ABAN

(0.2) 33 MP; [’Un]’e,

DIST.upriver.ALL


34 AB; [’un-], ABAN

35 is a little clos[er]. 36 MP; [ehn]-heh. 37 AB; Okay. 38 MP; ’Ungge,

DIST.upland.ALL

(0.3) 39 that’s all a,

(0.7) 40 all central,

‘ ‘To distantly upland’ is (also) central.’

41 AB; A:ll central, 42 okay.

(0.7) 43 MP; all,

{MP gestures to real Fairbanks}

91

44 I got the word [for you. 45 AB; [@@@@@] 46 MP; @] 47 That’s central.

(0.2) 48 AB; Central. 49 MP; Ehn-[heh]. 50 AB; [Ce--] 51 which one, 52 un-,

ABAN

(0.8) 53 ’un’e.

DIST.upriver.ALL

‘Which one is central, ‘to distantly upriver’?’

54 MP; ’Un’e, DIST.upriver.ALL

55 ’ungge, DIST.upland.ALL

‘ ‘To distantly upriver’ and ‘to distantly upland’.’

56 AB; okay, (1.8)

57 MP; ’unse is all ocean. DIST.ahead.ALL

‘ ‘To distantly ahead’ is all ocean.’

(0.4) 58 AB; okay. 59 MP; ’Unsghu.

DIST.ahead.AREA

‘A general area distantly ahead.’

(0.4) 60 AB; ’Unsghu.

DIST.ahead.AREA

‘A general area distantly ahead.’

61 MP; Ts’itu’, ((sic, intended: natu’)) major.river

‘Copper River.’

(0.4) 62 AB; Ts’it[u’].

major.river

‘Copper River.’

63 MP; [Na]tu’. ocean; salt

‘The ocean.’

{AB rearranges fabric roads to represent junction of Richardson and Glenn Highways}

92

64 AB; Natu’. ocean; salt

‘The ocean.’

65 MP; Ehn-[heh]. 66 AB; [So] this is Valdez, 67 down [2here, 68 MP; [2Ehn-heh2]. 69 AB; is that al2]so na[3tu’3]?.

ocean; salt

70 MP; [3Yeah3]. (0.2)

71 AB; Natu’?. ocean; salt

‘The ocean?’

72 MP; ’Unse, DIST.ahead.ALL

(0.5) 73 ’unse Bayliisde su kenii

DIST.ahead.ALL Valdez EMPH 3PL.SUB.say.IMPERF

de. SP/TEMP.ADVZ

‘ ‘Valdez is to distantly ahead’ they say.’

74 AB; ’Unse Bayliis[de. DIST.ahead.ALL Valdez

‘Valdez is to distantly ahead.’

75 MP; [Ehn-heh], 76 AB; Yeah]. 77 MP; Vald[2ez, 78 AB; [2Uh-huh2], 79 MP; they2] call it, 80 AB; [3Bayliisde3].

Valdez

81 MP; [3Bayliisde3]. Valdez

82 AB; Yeah, 83 MP; @@@@[@] 84 AB; [does] that mean something?.

(0.3) 85 MP; Ehn. 86 AB; That name, 87 Bayliisde,

Valdez

{AB gestures turning south on

Richardson Highway toward Valdez]

93

88 MP; Valdez. 89 it means Valdez.

(0.3) 90 AB; Sounds like that word, 91 yeah.

(0.5) 92 MP; Bayliis[de].

Valdez

93 AB; [Bay]liisde. Valdez

(0.3) 94 Okay. 95 MP; enmaec su kenii.

Fairbanks EMPH 3PL.SUB.say.IMPERF

‘They also say ‘ enmaec’ (i.e., for ‘Fairbanks’).

(0.4) 96 AB; enmaec?

Fairbanks

97 MP; enmaec [is Fair]banks. 98 AB; [ enmaec is he--]

Fairbanks

99 is here. (0.8)

100 Right?, 101 here’s Anch-, 102 Anchorage?, 103 MP; Ehn-heh,

(0.5) 104 AB; Vald[ez], 105 MP enmaec,

Fairbanks

106 right [2here, 107 AB; [2 enmaec2],

Fairbanks

108 MP; en2]maec. Fairbanks

109 AB; enmaec, Fairbanks

110 AB; up there. (0.5)

111 MP; <LOUD> enmae:c</LOUD>. Fairbanks

‘Fairbanks, Fairbanks, Fairbanks!’

{MP points erroneously at Anchorage; AB adjusts angle

of Richardson Hwy; AB points to Fairbanks at north end of Richardson Hwy}

{AB points at Anchorage}

{AB points at Valdez}

{AB and MP point at Fairbanks}

94

112 AB; Alright, 113 so how would you talk about,

(0.5) 114 ’un’e?,

DIST.upriver.ALL

‘To distantly upriver?’

(0.2) 115 MP; ’Un’e.

DIST.upriver.ALL


116 AB; ’un’e, DIST.upriver.ALL


117 and, 118 to Anchorage?, 119 ’unse?

DIST.ahead.ALL

‘To distantly ahead?’

(0.6) 120 MP; ’Unse.

DIST.ahead.ALL


(0.6) 121 AB; ’Unse.

DIST.ahead.ALL


(0.3) 122 MP; ’Unse.

DIST.ahead.ALL


(0.5) 123 AB; And that’s because it’[s, 124 MP; [Ehn-heh], 125 AB; on the w--] 126 MP; is all ocean. 127 AB; on the salt water, 128 [on the na]tu’.

ocean; salt

129 MP; [Yeah]. (0.3)

130 MP; Aen’, yes

131 natu’. ocean; salt

{AB gestures route to Fairbanks}

{AB gestures route to Anchorage}

{AB gestures route to Valdez}

{AB points to Anchorage and

Valdez simultaneously; MP gestures across both}

95

‘Yes, the ocean.’

132 AB; Natu’. ocean; salt

‘The ocean.’

133 Okay. ((Markle Pete, oai:paradisec.org.au:ALB01-057, 00:29:25.080-00:31:19.200))

In IU 04, Mr. Pete calls the direction to Anchorage ’unse’ ‘to distantly ahead’,

with an explanation about the proximity to salt water in IUs 08-19. In IUs 21-24, he

calls the direction to Fairbanks ’un’e ‘to distantly upriver’. In IUs 27-30 and 32-35 I

mention what I had recalled from the conversation in January 2010 (which is

documented in the fieldnotes shown in Figure 4.1), that ’unse refers to distances that

are far away, and that ’un’e refers to distances that are closer. However, in the

following IUs Mr. Pete does not abandon his claim that ’unse refers to locations

found on the marine coast (IUs 57-63). When I inquire about the proximity of

Valdez to salt water in IU 66, Mr. Pete confirms that the direction to Valdez would

be ’unse Bayliisde su kenii ‘Valdez is to distantly ahead they say’ in IUs 72-73. In

the following IUs Mr. Pete turns his attention back to Fairbanks (specifically, the

Ahtna name for Fairbanks, enmaec). In IUs 112-114 I ask him to confirm the

direction to Fairbanks as ’un’e ‘to distantly upriver’, which he had previously said in

IU 21. I then confirm his statement from IU 04 that Anchorage is in the ’unse ‘to

distantly ahead’ direction.

For just the three locations mentioned in (4.1), defining nse’ as referring to the

direction toward the ocean seems reasonable: Anchorage and Valdez are on salt

water, and Fairbanks is not. However, recall that Cordova, on the Gulf of Alaska, is

considered to be daa’ ‘downriver’ and is never referred to by Mr. Pete as nse’

96

‘ahead’. Thus ‘on a salt water coast’ is not a sufficient gloss for nse’. The four

locations by their proximity to salt water are shown in Table 4.3.

Table 4.3: Locations by nse’ and Proximity to Salt Water




no daa’

‘downriver’

on salt water no yes yes yes

4.2.1.3 Over “The Hump of the World”

A third intuition about the meaning of nse’ ‘ahead’ offered by Mr. Pete is related

to topography. In two separate sessions, he made reference to the idea that the earth

has a region of extreme elevation change that lies in proximity to the Copper River

valley and can be seen from certain key locations. He calls this area the “hump” or

the “ball” of the world, beyond which locations are in the nse’ ‘ahead’ direction.

The excerpt in (4.2) below occurred during our work with the diorama of the

trails around Tazlina Lake, depicted in the previous chapter in Figure 3.5. In IU 06

below Mr. Pete points out the change in direction that occurs when traveling

westward on the Glenn Highway: once past Tahneta Pass (IU 09; Map 3.1, G3), one

is traveling ’unse ‘to distantly ahead’.

(4.2) Nse’ ‘ahead’ beyond the “hump of the world” toward Anchorage 01 AB; He’s driving on the, 02 Glenn Highway? 03 MP; Car ’utgge natez’aan.

DIST.up.ALL 3S.SUB.clf.cpt.obj.IMPERF.ITER.INCEP

{AB moves toy car west on Glenn Highway}

97

‘The car starts to drive back to distantly up (vertically).’

04 AB; ’Utgge. DIST.up.ALL


(1.0) 05 MP; ’Unse [car na]tez’aan,

DIST.ahead.ALL 3S.SUB.clf.cpt.obj.IMPERF.ITER.INCEP

‘To distantly ahead the car starts to drive.’

06 AB; [’Unse]. DIST.ahead.ALL


07 MP; after you pass, (0.7)

08 the hill up there at um, (1.9)

09 Eureka. ((i.e. Tahneta Pass)) (0.4)

10 AB; A:h. 11 MP; Everything going down, 12 you see, 13 you see that. 14 AB; Yeah.

(0.8) 15 MP; Ehn-heh.

(0.6) 16 That’s the hump of the world, 17 I guess. 18 Yeah.

(0.5) 19 AB; Okay. 20 MP; Mm-hm. 21 AB; So everything past there, 22 once it starts to go back down, 23 MP; Down. 24 AB; towards Anchorage, 25 AB; it becomes, 26 MP; ’Utsene.

DIST.downland.ALL

‘To distantly downland.’

27 AB; it becomes ’unse. DIST.ahead.ALL

‘Once it starts to go back down it becomes ‘to distantly ahead’.’

28 MP; ’Unse or ’utsene. DIST.ahead.ALL DIST.downland.ALL

{MP gestures west on Glenn Highway}

{MP gestures toward real Eureka}

{MP gestures moving down a hill}

{MP gestures a vertical hump with both hands}

{AB gestures west of Tahneta Pass}

98

‘ ‘To distantly ahead’ or ‘to distantly downland’.’

29 AB; Either way, 30 okay. 31 MP; Either way, 32 you call them.


In IU 16 Mr. Pete describes the “hump of the world” (IU 16), which coincides

with Nekets’alyaexde ‘where we turn around’, the Ahtna name for the Tahneta Pass

region between the Copper and Matanuska drainages. Recall that Nekets’alyaexde is

an important location in Ahtna culture beyond which the declining terrain is salient

to all observers: “everything going down, you see, you see that” (IUs 11-13). This

steep descent is shown in the elevation profile of the Glenn Highway from the

Richardson Highway Junction to Anchorage in Figure 4.3; Tahneta Pass is the peak

at approximately mile 60.

Figure 4.3: Elevation Profile (Approximate) of the Glenn Highway

Mr. Pete again referenced the notion of a salient region of extreme elevation

change in a discussion of the difference between the direction to Valdez (nse’

99

‘ahead’) and the direction to Cordova (daa’ ‘downriver’). This discussion takes

place over a printed Ahtna region map, shown in Figure 4.4.5

Figure 4.4: Working with Ahtna Region Map

Of note in (4.3) is Mr. Pete’s recognition that the highway to Valdez takes its

directionals from the Copper River until Thompson Pass (Map 3.1, H5). Until this

point, traveling south on the highway is considered to be in the ’udaa’a ‘to distantly

downriver’ direction, but beyond this point the direction is considered to be ’unse ‘to

distantly ahead’. In IUs 01-03 I reconfirm that the direction to Valdez is ’unse ‘to

distantly ahead’ based on the prior conversation shown in IUs 65-73 of (4.1) above.

Mr. Pete agrees. I then reconfirm that the direction to Cordova is ’udaa’a ‘to

distantly downriver’, also based on our prior conversations (see Table 4.1 above).

5 Mr. Pete has is quite familiar with modern cartography from his many years

working in road construction.

100

(4.3) Nse’ ‘ahead’ beyond the “ball of the world” toward Valdez 01 AB; And then this one is,

(0.3) 02 ’unse.

DIST.ahead.ALL


(0.4) 03 Right? 04 MP; ’Unse.

DIST.ahead.ALL


05 AB, ’Unse. DIST.ahead.ALL


(0.5) 05 MP; Ehn-[heh]. 06 AB; [’U]daa’a.

DIST.downriver.ALL


(1.8) 07 MP; They call it ’udaa’a halfway to Valdez,

DIST.downriver.ALL

‘They call it ‘to distantly downriver’ halfway to Valdez,’

08 where Valdez at #, (0.8)

09 oh Val@dez, 10 AB; @@. 11 MP; ’Udaa’a,

DIST.downriver.ALL

‘ ‘to distantly downriver’,’

(0.2) 12 AB; Uh-huh, 13 MP; and then you, 14 ’unse.

DIST.ahead.ALL

15 When they get to Thompson Pass.

‘and then you (go) ‘to distantly ahead’ when you get to Thompson Pass.’

{AB gestures route south from Glennallen to Valdez}

{AB gestures route south from Glennallen to Cordova}

{AB and MP point at Valdez}

{MP gestures “turning corner” on route to Valdez”

{MP points at Thompson Pass}

101

16 AB; Oh like, 17 so,

(0.7) 18 MP; ’Utsene.

DIST.downland.ALL

‘To distantly downland.’

19 AB; mm-hm, (0.2)

20 MP; ’Unse, DIST.ahead.ALL

21 MP; they [say]. ‘ ‘To distantly ahead’ they say.’

21 AB; [’un]se. DIST.ahead.ALL


(0.5) 22 AB; Once you start, 23 getting over the pass, 24 and going back [down. 25 MP; [Ehn-heh, 26 AB; That’s ’unse].

DIST.ahead.ALL

27 MP; dropping]. (0.6)

28 AB; Okay. (0.6)

29 MP; That’s where the, 30 ball at [with], 31 AB; [Yeah].

(0.9) 32 MP; with the world, 33 see,

(1.4) 34 AB; The top. 35 MP; Ehn-heh.


As in (4.2) above, there is a change of directional to nse’ ‘ahead’ once past the

summit of a mountain pass. Mr. Pete acknowledges in IUs 27-34 that the salient

factor triggering nse’ ‘ahead’ is the steep decrease in elevation beyond Thompson

Pass, which he refers to as the “ball” of the world. Indeed the decrease in elevation

102

between the top of the pass and Valdez is rapid, shown in the elevation profile of the

Richardson Highway between the Glenn Highway junction and Valdez in Figure 4.5.

The highway descends from 2805 ft. to nearly sea level in approximately thirteen

miles.

Figure 4.5: Elevation Profile (Approximate) of the Richardson Highway

From the Glenn Highway Junction to Valdez

Assuming for now that nse’ ‘ahead’ refers to locations beyond the summit of a

high pass, let us examine Cordova and Fairbanks, the two locations not mentioned in

(4.2) or (4.3). Mr. Pete’s insistence that Cordova is not nse’ ‘ahead’, but rather daa’

‘downriver’ is in keeping with the notion that nse’ refers to locations beyond a steep

mountain pass. The route to Cordova is riverine, and one need not surmount a pass in

order to get there, as noted above, one needs to either take a boat or, in winter, a sled.

Fairbanks, on the other hand, is beyond a steep mountain pass. Isabel Pass

through the Alaska Range (Map 3.1, D5) is at an elevation of 3258 feet, higher than

either Tahneta or Thompson Passes, yet locations beyond it are not considered to be

in the nse’ ‘ahead’ direction. Thus ‘beyond a steep mountain pass’ is also not a

103

sufficient gloss for nse’. Table 4.4 shows the four locations by their relation to a

mountain pass.

Table 4.4: Locations by nse’ and the Need to Cross a Mountain Pass




no daa’

‘downriver’

beyond a mountain

pass yes yes yes no

4.2.2 Other Possibilities From the Literature

Now that we have seen that Mr. Pete’s three intuitions on the meaning of nse’ –

‘far away’, ‘on salt water’, and ‘beyond a pass’ – are individually not sufficient to

account for nse’ being appropriate for Anchorage and Valdez, but not for Fairbanks

and Cordova, let us examine two possible analyses of the semantics of nse’ that

could be gleaned from the discussion in Kari (2010).

4.2.2.1 Association With the Arc of the Sun as a Metaphor for the

Hearth

First, Kari (2010) draws an analogy between the indoor use of nse’ to refer to the

direction from the back wall of the living area toward the hearth in the center of the

room on the one hand, and the outdoor use of nse’ to refer to the arc of the sun

across the southern sky on the other: “it appears that [the semantics of nse’ on the

large, geographic scale] is based upon ‘in front, toward the hearth’ in the sense that

104

one is facing the ‘hearth’ as the arc of the sun in the southerly sky” (2010:133).6 At

such high latitudes, however, the arc of the sun varies widely over the year, covering

an azimuth angle between 145° and 215° at the winter solstice, and between 30° and

330° at the summer solstice as shown in Figure 4.6.7 While Kari only posits this

connection as an intriguing historical path of semantic extension between the indoor

and outdoor uses of nse’, it is nonetheless not a sufficient gloss for the synchronic

meaning of the directional, as different locations would be subsumed under the arc

of the sun at different times of the year.8

6 Kari further speculates on the ‘hearth’ sense of the term: “One faces a fire, so it

is in front, whether indoors or outdoors. Thus it is autonomous from the four-part riverine set of roots. One moves ‘forward’ as instinctually moving the body toward a fire” (p.c.).

7 The sun’s azimuth angle is defined as the direction of the sun in the horizontal

plane from a given location, with north defined as having an azimuth of 0° and south as having an azimuth of 180°. For our purposes, this means that from Copper Center, the distance between the locations on the horizon of sunrise and sunset is narrow at the winter solstice (about 70°) and wide at the summer solstice (about 300°).

8 Kari notes: “[Alaskan Athabascan languages] Gwich’in, Lower Tanana,

Dena’ina, Ahtna, and Tanacross all use some version of nse’ for the general direction of an arc to the south, [but] Koyukon may have a somewhat different use of ne e [the Koyukon cognate]. Koyukon does not use ne e for fire direction, cooking pit, etc. […] Lower Tanana seems to have the most conservative use of [the cognate] ntha’ ‘to the fire’” (p.c.).

105

Figure 4.6: Path of the Sun as Seen From Copper Center, December - June9

4.2.2.2 A Lexicalized Toponym

A second interpretation to be gleaned from Kari is based on the Ahtna

extraterritorial place name Danse ‘to near ahead’. Kari (2010) interprets Danse as

referring to the area beyond the Chugach range, which includes Cook Inlet and

Anchorage to the west-southwest, and Prince William Sound and Valdez to the

south. Note that the area designated by this place name stands “opposite” another

extraterritorial place, also beyond a major mountain range, for which the toponym

has lexicalized from a directional stem. Ba’aaxe ‘an area outside’, based on the stem

9 Created using the Sun Path Chart Program, University of Oregon Solar

Radiation Monitoring Laboratory. http://solardat.uoregon.edu/SunChartProgram.php.

106

’an ‘outside’, is the name of the greater Tanana River valley to the north of the

Alaska Range. This might suggest that the synchronic meaning of the directional

nse’ is taken from the regional toponym to refer to directions toward that area alone.

4.3 Toward a Unified Definition of nse

Table 4.5 is an aggregate table of the possible semantics of nse’ presented in this

chapter. For the sake of completeness, ‘downriver’ has been added as well, since two

of the towns are located at the mouths of rivers. Leer (1989) points out that ‘ahead’

and ‘downriver’ as associated semantically if one thinks of rivers as flowing ‘ahead’.

As we can see, no single feature here fully accounts for the attribution of nse’ to

Anchorage and Valdez, but not to Fairbanks and Cordova. Instead, the term is best

classified as a network of related meanings, with a few features more prominent than

others as core meanings. There are two possible combinations of core meanings for

nse’: first, that nse’ refers to locations that are both (a) on salt water and (b) require

traversing a steep mountain pass; or second, that nse’ refers to locations that are (a)

beyond the Chugach Mountains specifically, and (b) require traversing a steep pass

mountain pass.

107

Table 4.5: Aggregated Possible Meanings of nse’




no daa’

‘downriver’

at a “far” distance

yes no no yes

on salt water no yes yes yes

beyond a mountain

pass yes yes yes no

“southern hearth”

no ?

(WSW) yes yes

beyond Chugach

Mountains no yes yes yes

downriver no yes ?

(route partly down river)

yes

The salt water requirement is one way that Fairbanks is eliminated as a potential

nse’ location, but this implies that there is some cognitive import in differentiating

the ocean’s edge within the directional system. This may in fact be the case for the

Ahtna, and for nearly all Northern Athabascans. The notion of ‘perimeter’ or ‘edge’

is part of the semantics of reflexes of Proto-Athabascan *n s- ‘ahead; out into open

water’ in at least Koyukon (Jetté & Jones 2000:467) and Dena’ina (Kari 2007:336ff).

In fact most Northern Athabascans, having likely come from an urheimat much

further inland (Kari 1996a) have never been particularly oriented toward the ocean

108

for subsistence, and the single group to now occupy territory on a salt water coast,

the Dena’ina, were lacking not only terminology for many aspects of maritime life,

but also the technology for harvesting marine resources. The Ahtna have

traditionally focussed on acquiring resources from rivers and inland game. In this

sense, the salt water may be perceived of as an edge or barrier it is not possible to

cross, which opens the door for the notion becoming lexicalized in the Ahtna

directional system.

Fairbanks is alternately eliminated by the feature that associates nse’ with areas

beyond the Chugach Mountains, because of an association with the regional place

name Danse ‘to near ahead’. Identifying an individual mountain range may seem

overly particular for a directional system until one recalls that the opposite direction

– locations beyond the Alaska Range in the Tanana River valley – are traditionally

referred to by a similar construction, the regional place name Ba’aaxe ‘an area

outside’. Thus there seems to be a semantic and geographic opposition between the

mountain range to the north Ahtna territory and the one to the south, which Kari

acknowledges (2010:12).10

In both possible combinations of features, Cordova is

eliminated from the list of nse’ locations by the topographical requirement that such

locations be beyond a mountain pass.

Let us return to Kari’s interpretation of nse’ as referring only to the locations

beyond the Chugach Range. There is one sense in which it might be possible to see

10 We have seen from Chapter 3 that Mr. Pete does not have the association

between the direction to Fairbanks and the ’an (or ’ane’) ‘outside’ directional stem (although the lexicalized place name Ba’aaxe is still active in his lexicon).

109

this as the only required feature of nse’. If the Ahtna conception of the entirety of the

Chugach Mountains does not include the eastern section of the range through which

the Copper River passes to reach the Gulf, then nse’ would naturally only include

Anchorage and Valdez. The northwest-southeast arc of the range puts the eastern

sections out of visual distance from the Copper River valley at least, and it is entirely

possible that the US Geological Survey’s definition of the Chugach Mountains

includes a far wider span than the Ahtna cognition of the area does. However, I

doubt this is the case. The Ahtna, like all Athabascans, were experienced travelers

(Kari 2008, 2010) and would surely have been aware that there is no physical

division between the range one must traverse to get to Anchorage and Valdez on the

one hand, and the range the Copper River cuts through toward Cordova on the other.

Instead I believe that as PA *n s- ‘ahead; out into open water’ developed into the

Ahtna cognate nse’, the notion of the steep mountain pass came to be a salient

feature of the semantics of the term. This development is particular to the geography

of Ahtna territory, in which two routes toward open water require traversing a steep

mountain pass (Tahneta Pass toward Cook Inlet, and Thompson Pass toward Prince

William Sound). As was mentioned in chapter 3, mountain passes present obstacles

that must be traversed with a great deal of effort, and the technologies and resources

that are required to cross a mountain range are very different from those needed to

navigate a river. Even the seasonal cycle in the far north becomes a major issue. In

winter, Thompson Pass was not reasonably traversable due to extremely heavy

snowfall, and a traveler heading to the salt water coast would instead have opted for

a journey atop the frozen river via dogsled. In summer the Copper is only navigable

110

by boat, and foot travel through the pass is possible. Thus distinguishing between the

nse’ path to the sea and the daa’ path to the sea becomes very important in semi-

nomadic culture. The “hump of the world”, then, is the portion of the Chugach

Mountains that stands as a barrier between the Ahtna region and extraterritorial areas

that are either on salt water or simply opposite Ba’aaxe.

4.4 Conclusion

In this chapter I have examined five separate possibilities for refining the

meaning of the directional nse’, which has traditionally been glossed as ‘ahead’.

While my consultant was never uncertain about which locations we discussed were

in the nse’ direction and which were not, his intuition presented us with options,

none of which alone fully accounts for the data he provided on its own. His

suggestion of ‘far away’ does not account for the proximity of Valdez, which is nse’,

nor for the long distance to Fairbanks, which is not nse’. His suggestion of ‘on a salt

water coast’ does not account for Cordova not being nse’. Finally, his suggestion of

‘beyond a steep mountain pass’ accounts for Anchorage and Valdez being nse’ and

for Cordova not being nse’, but it misses the mark for Fairbanks, which is over a

steep pass and is not nse’.

I also discussed two possibilities arising from the discussion of nse’ that Kari

(2010) provides. First, his assertion that nse’ refers to the arc of the sun across the

sky as an analogy for the indoor use of nse’ to refer to the hearth-ward direction,

while likely a path of historical semantic extension between the two, does not

provide an adequate gloss for the directional. This is because the path of the sun

111

varies widely throughout the year, and locations subsumed under the arc would

change from month to month. A second possibility arising from Kari’s work is that

nse’ refers to locations beyond the Chugach Mountains, but this does not account for

Cordova, which is also beyond that mountain range.

It seems that nse’ is best described as a network of related semantic features,

some arising from historical meanings of the term, and others arising diachronically

in the Ahtna territory. A few semantic features seem to have developed into core

senses of the term. These are either ‘over a steep pass and to the edge of the ocean’,

or ‘over a steep pass through the Chugach’. Whichever of these two combinations is

the synchronic semantic basis of nse’, the salient feature here is the notion of a steep

pass. Again, topography is salient in the Ahtna directional system. Foot travel

through mountainous areas requires a different kind of effort from winter or summer

travel on a waterway. As with the notion of of ngge’ ‘upland’ discussed in Chapter 3,

the acknowledgment of elevation is part of the directional paradigm of Ahtna. The

paradigm is more three-dimensional than the English system, which only

acknowledges differences in one plane.

Some questions about the semantics of nse’ remain, of course. First, how would

one gloss this term? Second, how typologically likely is this meaning? Third, do

speakers make a distinction between a local description of direction and a regional

one? Note that in both (4.2) and (4.3) above ’utsene ‘to distantly downland’ would

have been an acceptable description of the downhill paths toward Anchorage and

Valdez. Calling these routes ’utsene is descriptive from a local point of view, while

112

calling them nse’ shows an awareness of the routes as arteries connecting larger

regions.

113

Chapter 5

Directional Reference in Discourse and Narrative:

Comparing Indigenous and Non-indigenous Genres

The previous chapters deal primarily with Ahtna directional reference in

isolation, that is, with the directional system as it is revealed in elicitation in a setting

designed to focus attention on this part of the grammar. While elicitation is

invaluable for understanding the subtleties of and changes to the synchronic

semantics of the directionals, it does not allow us to observe how speakers “do”

directional reference in spontaneous, connected speech. Let us now turn to exactly

that. The remainder of this dissertation deals with Ahtna directional reference in

narrative genres of discourse.

The present chapter situates the study of the expression of direction in Ahtna in

the recent typological literature based on “frog story” studies, a popular fieldwork

methodology designed to elicit samples of connected speech that are comparable

across languages. This chapter presents evidence from an Ahtna speaker’s telling of

a frog story that shows that for Ahtna, the genre does not fully exemplify the

linguistic resources for describing direction that are available to speakers. Instead we

need to turn to an indigenous genre of discourse known as travel narratives, in which

speakers fully exploit a range of grammatical systems to present vivid descriptions

of direction and location. Furthermore, this difference arises from a fundamental

contrast between the discourse structure of frog stories and travel narratives. Travel

114

narrators use discourse particles to construct vivid descriptions of the geographic

ground against which the figures move, while the frog-story narrator is instead

concerned with the figures themselves, using complex clauses to track characters

through the narrative while remaining relatively vague about the landscape they

move through.

This chapter sets the stage for the following chapters, which examine Ahtna

travel narratives more closely in terms of how speakers employ the complex

morphology of the directionals in spontaneous speech.

5.1 Frog Stories and Travel Narratives in the Study of Direction

and Location

This chapter is an overview of Ahtna speakers’ attention to spatial concepts in

spontaneous connected speech, presented against a background of typological

research into the crosslinguistic patterns for expressing direction and location in

language. Since the mid-1990s, linguistics has seen a proliferation of literature based

on “frog story” experiments. In this research paradigm, speakers of various

languages are shown a textless children’s book of drawings called Frog, Where Are

You? (Meyer 1969) and are asked to narrate the events depicted there in their native

tongue.1 The story is about a boy and his dog who search through the woods for a pet

1 Frog story research developed out of the desire to make typological

comparisons across languages. Because direct translations from a contact language can often distort the native grammar, one solution has been to present speakers of different languages with the same nonlinguistic stimulus in order to elicit

115

frog that had escaped from its glass jar during the night. The boy and the dog

undergo a series of adventures in which they encounter a host of forest dwelling

creatures, including a squirrel, a swarm of angry bees, an owl, and a buck. The

characters move from place to place, starting with a fall from the bedroom window

and ending climactically with a ride atop the buck’s horns and a tumble from a cliff

into a river.

Over the years, narratives resulting from frog-story experiments have been used

as the basis for inquiry into a wide range of topics in linguistics and psychology.

Among these are:

• the acquisition of grammatical and narrative structure in L1 (e.g., Berman

and Slobin 1994; Orsolini et al. 1996; Hoff-Ginsberg 1997; Bavin 1998,

2000; Morgan 2002; Boumans 2006; Winskel 2007) and L2 (e.g., Kang

2004; Chini and Lenart 2008),

• the effects of bilingualism on narrative structure (e.g., Bennet-Kastor

2002; Morford 2002; Ordoñez 2004; Verhoeven 2004; Serratrice 2007),

and

• a typological perspective on how different languages exploit lexical

structure to express direction (or “path”) and manner in motion events

(e.g., Slobin 1996; Bavin 2004; Brown 2004; Engberg-Pederson and

Trondhjem 2004; Galvan and Taub 2004; Ibarretxe-Antuñano 2004a,

spontaneous speech, which is more likely to reveal authentic grammatical constructions.

116

2004b; Jovanovic and Martinovic-Zic 2004; Ragnarsdóttir and

Strömqvist 2004; Slobin 2004; Zlatev and Yangklang 2004).

The last of these topics – the study of the expression of direction or “path”

description in language – has garnered much attention. The research, summarized in

Slobin (2004), stems from the work of Talmy (1985, 1991, 2000), who examines the

lexicalization of semantic units and the patterns of conflation of those units (i.e.,

MOTION + MANNER, MOTION + PATH, MOTION + FIGURE). Slobin (2004) presents a

typology in which languages can be characterized as belonging to one of three types.

“Verb-framed languages” are those in which the description of the path or direction

of a motion event is encoded in the verb. “Satellite-framed languages” are those that

encode direction in a satellite element, usually some kind of adjunct like an

adpositional phrase or an adverb.2 “Equipollently-framed languages” are those in

which manner and direction are expressed by elements of equal status, e.g., serial-

verb languages.3 Strömqvist and Verhoeven (2004) is a collection of studies aiming

2 Many readers will recognize the commonly cited illustration of verb-framed

and satellite-framed languages: Spanish, a verb-framed language, encodes path in the verb salió ‘exited’ in la botella salió de la cueva flotando, while English, a satellite-framed language, encodes path in the satellite out in the bottle floated out of the cave.

3 Slobin gives a tantalizing but otherwise unsupported p.c. from Richard Rhodes

regarding equipollently-framed languages: “[…] such constructions are typical of the American Indian languages Algonquian, Athabascan, Hokan and Klamath-Takelman” (Slobin 2004:247, emphasis added). Berez (2009) presents arguments against Athabascan languages fitting into the current three-way typology at all, let alone as an example of an equipollently-framed language. The reasons include definitional ambiguity for the notion of “satellite” and Athabascan-specific issues with defining “word-hood” and “equal status”. This issue will not be taken up further here.

117

to apply this typology to Warlpiri, Basque, Tzeltal, Icelandic, West Greenlandic,

Swedish, Thai, American Sign Language, and Arrernte.

Among these, Wilkins (2004) in particular finds that extralinguistic ethnographic

considerations are very closely tied to grammar in his study of frog stories in

Arrernte, a language of Central Australia. He shows that even within the genre of

frog stories, Arrernte culture is a better predictor than language type of motion event

segmentation. Arrernte is, according to Slobin’s typology, a verb-framed language,

so one could hypothesize that Arrernte speakers devote less attention to the dynamic

description of direction than would speakers of satellite-framed languages (this is

one of the predictions of the typology; see Slobin 2004). Conversely, though, based

on the nomadic culture of Desert Aborigines and the prominence of travel in

everyday life, one could also hypothesize that Arrernte speakers will pay special

attention to routes of motion and should construct elaborately detailed direction

descriptions (see Wilkins 2004 for the argumentation). Wilkins shows that the latter

is true: Arrernte speakers segment the direction description into more distinct

trajectories in frog stories than English speakers. The difference is both quantitative

in terms of path complexity and qualitative in terms of the kinds of linguistic

structures used by speakers. As Wilkins writes, “[t]hus, it is the areal ethnographic

observations … which here appear to be more predictive of the findings [i.e., than

the typological predictions]” (2004:155). The role of culture on the development of

grammar should not be underestimated.

This chapter examines the role that cultural considerations play in the grammar

of motion events, and in the efficacy of frog stories in eliciting that grammar, in

118

Ahtna. Like the Desert Aborigines, Ahtna society is traditionally semi-nomadic. As

is mentioned in Chapter 1, hunters and family groups traveled seasonally in pursuit

of resources like fish and big game (Reckord 1979, 1983a, 1983b). Knowledge of

the surrounding terrain is not only essential to survival but also plays an important

role in ethnic identity and the assertion of the connection of one’s social group (tribe,

band, family) to the land, much like Moore and Tlen (2007) found for Athabascan

speakers in the Yukon. Individual Ahtna men – and some women – are often

intimately familiar with large swaths of the 35,000 square miles of Ahtna territory

and beyond, a feat all the more impressive for having been undertaken on foot or by

dogsled.

The importance of geographic knowledge and ‘travel talk’ is reflected in the

sheer size of the corpus of Ahtna place names (Kari 1983, 2008). The corpus

contains over 2,200 names, many of which are documented in a genre of oral

literature that Kari terms elite travel narratives. These narratives are a kind of

“virtual guided tour” in which the speaker discusses, in sequential order, all the

meaningful and hence named locations along a given route. A single narrative may

cover over one hundred miles of river and/or trail and is often interspersed with

personal memories and descriptions of how each site was used seasonally for

camping and hunting (for published examples of Ahtna travel narratives see Kari

1986; Kari & Fall 2003; Kari 2010).

In some superficial ways, frog stories and Ahtna travel narratives are similar:

animate referents move across the countryside in pursuit of animal(s). But in many

ways, particularly cultural ways, they are different. In Frog, Where Are You?,

119

referents engage in activities that do not happen everyday: heads get stuck in jars,

characters fall out of windows and trees and off cliffs, owls and gophers pop

suddenly out of their holes, and characters interact with rather unfriendly bucks and

bees. In travel narratives, on the other hand, activities are generally limited to

walking, sledding, hunting, and camping.

As is discussed below, speakers telling travel narratives make full use of the

grammar of path and location available to them, including adverbial verb prefixes, a

class of riverine directionals, and highly systematic toponymy. Interestingly, while

all of these are also available to frog-story narrators, speakers in this genre seem to

restrict themselves to only a narrow range.

What role, then, does genre (e.g. Mayes 2003) play in an academic study of how

a language encodes notions of direction and location in motion events? Can a frog

story fully reveal the nature of Ahtna grammar about motion? Or will other

concerns, specific to the tasks of telling a frog story or a travel narrative, be more

important and ultimately influence where a speaker’s attention lies? The following

sections examine two Ahtna travel narratives and an Ahtna frog story with an eye

toward answering these questions.

The first travel narrative was recorded in 1980 by Jake Tansy with linguist James

Kari. Mr. Tansy describes an overland and riverine hunting route, used exclusively

in the summertime, from the mouth of Alaska’s Brushkana River to the Yanert Fork

and onward to Valdez Creek. In Ahtna the story is called Saen tah xay tah c’a

u’sghide ‘we used to travel around in summer and winter’ (Kari 2010: 59-69);

120

henceforth Mr. Tansy’s monologue is referred to as Saen tah xay tah xay tah

(‘during summer and winter’).

The second narrative was recorded by Adam Sanford in 1986, also with Kari.

This is an epic description of yearly hunting routes, often in the extreme

mountainous highlands in pursuit of Dall sheep. The entire recording is nearly thirty

minutes in length; only the first five minutes are presented here. This narrative is

known in Ahtna as C’uka ts’ulaen’i gha nen’ ta’stede dze’ ‘how we went hunting

out in the country’ (Kari 2010:91-128). Henceforth this narrative is referred to as

Ta’stede dze’ (‘we went hunting thus’). Finally, the narration of Frog, Where Are

You? – in Ahtna Naghaay, ndaane zidaa, and henceforth Naghaay (‘frog’) – was

recorded in October 2008 in Tazlina, Alaska by speaker Markle Pete with the author.

Section 2 below compares the distribution of the spatially oriented grammatical

systems across the two genres. I first look at the use of direction- and location-

describing adverbial verb prefixes, a mechanism that is employed by all three

speakers. I then look at two areas of the grammar where the stories differ, the use of

directionals and toponymy. While all three systems are fully utilized by travel

narrators, Mr. Pete makes very limited use of the last two in Naghaay.

The unequal use of spatial grammar in the two genres reflects the speakers’

unequal attention to figure and ground. Section 3 examines how the storytelling tasks

that the speakers deem important influence the choices they make when structuring

discourse. Mr. Tansy and Mr. Sanford foreground the spatial and temporal

trajectories of their narratives, while Mr. Pete elects instead to carefully track

referents, which is common in frog-story narration worldwide.

121

Section 4 contains concluding remarks about the role of genre in typology and

language documentation. Ultimately it seems Mr. Pete’s concerns may not lie in

creating a fully elaborated sense of the fictional landscape in Frog, Where Are You?,

which in turn affects how well the story reveals Ahtna’s path-describing grammatical

systems. As is discussed below, works of oral literature in Ahtna are often a-spatial.

5.2 Grammar of Direction and Location in Saen Tah Xay Tah,

Ta stede Dze , and Naghaay

This section describes three mechanisms for elaborating direction and location in

Ahtna: adverbial verb prefixes, riverine directionals, and toponymy. It compares the

travel narratives to the frog story in terms of each of these linguistic resources.

Before continuing, however, it is crucial to clarify the distinction between the

Ahtna lexical class of directionals and the semantic category of direction. The

former refers to the lexical class that was presented in Chapter 2 and discussed in

Chapters 3 and 4, the class of Ahtna words that are behaviorally united by the

tripartite morphological structure described in Table 2.1 and have a largely, though

as we have seen not total, riverine basis. The latter, however, is a larger semantic

category that is found across languages and in which several areas of Ahtna grammar

participate. In Ahtna, adverbial verb prefixes, postpositions, demonstrative pronouns,

and directionals are all potentially available to speakers for the function of

describing direction. For the purposes of this dissertation, members of the Ahtna-

specific lexical class will be referred to as directionals while the crosslinguistic

semantic category will be referred to as direction. This distinction is especially

122

important to a discussion of frog story research, for which the goal has been to

develop a crosslinguistic typology of direction.

5.2.1 Derivational/Thematic Adverbial Prefixes

Recall from Chapter 1 that Ahtna, like all Athabascan languages, is a

polysynthetic language with templatic verbal morphology. Table 5.1 shows a

simplified version of the verb template from Kari (1990). Verbs are stem-final, with

eleven prefix slots to the left of the stem. Near the far left edge, in position ten, we

find the so-called ‘derivational/thematic’ prefixes. Many of the morphemes here are

adverbial in function and can describe, among other things, direction and location.

Kari writes of the morphemes found here, “nearly one hundred morphemes appear in

this position … includ[ing] bound postpositions” (1990:40), which provides insight

into the source of the spatial nature of these prefixes. The morphemes in this position

that were historically free preverbal postpositions are grammaticalizing; they are

undergoing phonological fusion to the verb, losing their objects, and becoming less

adpositional and more adverbial in function.

Table 5.1: Ahtna Verb Template (adapted from Kari 1990)

Bo

un

d p

ost

po

siti

on

al

obje

ct

Der

ivat

ional

/ T

hem

atic

Iter

ativ

e

Dis

trib

uti

ve

Inco

rpora

te

Them

atic

Pro

no

min

al

Qu

alif

iers

Conju

gat

ion

Su

bje

ct

Cla

ssif

ier

Ste

m

11 10 9 8 7 6 5 4 3 2 1

123

In Saen tah xay tah, Ta’stede dze’, and Naghaay, all three speakers make

extensive use of direction-describing adverbial prefixes. Of the seventy-eight motion

verbs in the three stories, only four have no such prefix. Furthermore, the prefixes

occur with uniform density. In Saen tah xay tah twenty-one of twenty-four motion

verbs contain at least one direction- or location-describing prefix; the ratio in

Ta’stede dze’ is twenty-one of twenty-two motion verbs; and in Naghaay, thirty-two

of thirty-two motion verbs. These distributions do not differ significantly.

In contrast to the highly precise directional system, the adverbial prefixes usually

describe simple paths of motion of the subject. In the examples from Saen tah xay

tah and Ta’stede dze’ shown in (5.1-5.2), the prefixes encode the simple directions

of ‘away’, ‘out’, ‘through’, ‘back’, and ‘up’.

(5.1) Direction-describing adverbial verb prefixes in Saen tah xay tah 01 JT; Xona,

now

(0.6) 02 first

(0.9) 03 nen’ ta’stghide de c’a saen ta,

country 1PL.SUB.pl.go.away FOC summer during

(0.9) 04 c’a Bes Ggeze Na’,

FOC bank worn stream.POS

(0.4) 05 Saas Nelbaay Na’,

sand grey stream.POS

06 hwcets’ede . 1PL.SUB.ascend

‘When we first went out in the country during the summer we would go to the base of ‘Worn Bank Stream’ or ‘Sand That is Grey Stream’.’

(1.5) 07 Ni denta u,

sometimes EVID

124

(0.6) 08 Dghateni yi ’e tanidzeh,

stumbling.trail 3s CONJ middle.water

09 xu Dghateni ts’idini en. stumbling.trail water.flows.out

‘Sometimes also to ‘Stumbling Trail’ or the middle one flowing out from ‘Stumbling Trail’.’

[…] 43 dets’en,

next

44 Nts’ezi Na’ ba’aa, N. stream.POS outside

45 dghilaay ghakudaan de kanats’ede mountain hole.extends.through 1PL.SUB.pl.go.ITER.up n’e ,

CONJ

‘outside of ‘Nts’ezi’s Stream’ we went back up where a tunnel extends through the mountain, (and…)’

((Jake Tansy, Saen Tah Xay Tah C’a u’sghide ‘We Used to Travel Around in Summer and Winter’,

00:00:05.580-00:01:20.650. Kari 2010:60-61))

(5.2) Direction-describing adverbial verb prefixes in Ta’stede dze’ 127 AS; K’a xona yet hwts’en xona na’stetnaesi,

then there from.area then 1PL.travel.nomadically.back

(2.3) 128 ohh dahwtne dak.

oh 3S.be.steep

‘Then as we moved back from there, oh it (the canyon) was steep.’

(0.6) 129 Ni k’aedze’ dahwtne dak xona,

both.sides 3S.be.steep then

(0.8) 130 saanetah kats’enaes.

barely 1PL.trave.nomadically.up

‘It was steep on both sides and then we could barely move up.’

((Adam Sanford, C’uka Ts’ul’aen’i gha Nen’ Ta’stede dze’

‘How We Went Hunting Out in the Country’, 00:03:56.740-00:04:06.320. Kari 2010:96))

The semantic generality of the prefixes allows them to be used in a range of

situations, not only describing a journey across the countryside as in (5.1-5.2), but

125

also for the up-and-out motion of a squirrel from his den and the downward

orientation of a bees’ nest hanging from a tree branch, as in the excerpt from

Naghaay in (5.3).

(5.3) Direction-describing adverbial verb prefixes in Naghaay 13 MP; ic’ae gilok'ae naghalts’et.

dog window 3S.falls.down

‘The dog falls down from the window.’

((Markle Pete, Naghaay Ndaane Zidaa ‘Frog Where Are You’,

oai.paradisec.org.au:ALB01-AHTNB001-059.tiff))

[…] 22 Dligi kaniyaa,

squirrel 3S.go.up.and.out

‘A squirrel comes up,’

23 ic’ae ngga t’ox naggic’e ’i gha itsae. dog upland nest 3S.hang.down.REL at 3s.bark

‘the dog barks at the nest that is hanging (there).’

((Markle Pete, Naghaay Ndaane Zidaa ‘Frog Where Are You’, oai.paradisec.org.au:ALB01-AHTNB001-060.tiff,

ALB01-AHTNB001-061.tiff))

Based on the uniform density of spatially oriented prefixes in all three narratives,

it would be reasonable to claim that frog story experiments do indeed exemplify how

Ahtna speakers use prefixes to describe direction concepts in discourse. However,

other data shows that Naghaay does not provide us with a complete picture of how

richly speakers can describe direction and location in Ahtna. The next section

discusses the use of the directional system that is employed extensively in the travel

narratives, and only minimally in Naghaay.4

4 Mr. Pete makes only limited use of the grammar of path and location because

he is attending to other concerns, not because he is unfamiliar with these portions of

126

5.2.2 Directionals in Discourse

As was shown in Chapters 2-4, the morphological structure of the lexical class of

directionals potentially allows speakers to be very precise in describing paths and

locations in terms of their relationship to the placement and flow of the local river. In

addition, the structure of Ahtna discourse allows even more specificity. Speakers

very often use multiple directionals in a single clause to describe changes in

trajectory or to pinpoint a destination precisely. In (5.4), Mr. Tansy uses multiple

directionals, in addition to adverbial prefixes, to describe a complex path with

several trajectory changes (downriver – across – downriver – back).

(5.4) Use of directionals in Saen tah xay tah 25 JT; Ni denta u’,

sometimes EVID

(0.7) 26 yet,

there

(0.8) 27 Tl’ahwdicaax Na’,

headwaters.be.valuable stream.POS

(0.4) 28 ’udaa’a,

distant-downriver

(0.5) 29 ’unaa daa’a ts’its’ede dze’ dae’,

distant-across downriver 1PL.go.out thus

30 Nts’ezi Na’ hwts’e’, N. stream.POS from.area

the grammar or an unskilled storyteller. Like all speakers, Mr. Pete uses the discourse mechanisms that are relevant to the task at hand, as is shown in section 0.

127

(0.6) 31 tes ninats’ede .

pass 1PL.go.back.to.a.point

‘Sometimes then, we come out downstream and across and downstream of ‘Valuable Headwaters Stream’ and we come back to a pass at ‘Nts’ezi’s Stream’.’


00:00:46.710-00:00:57.290. Kari 2010:61))

Saen tah xay tah contains twenty-two directionals over 100, and Ta’stede dze’

contains twenty-six directionals over 218 IUs. Naghaay, however, contains only five

(over 55 normative sentences; IUs were not available). Of the handful of directionals

found in Naghaay, three occurrences are the use of ngge’ ‘upland’ to describe the

location of the bees’ nest. The nest is not actually upland from a river – in these

pages there is no river to be seen – but the term is used idiomatically to mean ‘over

there’ or ‘up that-a-way’. Mr. Pete translated the line in (5.5) with a small backhand

wave gesture:

(5.5) Use of ngge’ ‘upland’ in Naghaay 19 MP; Ngge’ t’ox naggic’e ’.

upland nest 3s.hang.down

‘Up (over there) the nest is hanging.’

((Markle Pete, Naghaay Ndaane Zidaa ‘Frog Where Are You’, oai.paradisec.org.au:ALB01-AHTNB001-060.tiff))

In (5.6) the nonriverine directional ’utgge’ ‘distantly up vertically’ is used to

describe the owl’s perch.

(5.6) Use of ’utgge’ ‘distantly up vertically’ in Naghaay 39 MP; Besiini ’utgge’ dazdaa.

owl DIST.up.vertically 3s.sit

“The owl is sitting up there (i.e., high up, on a branch).’


128

The final directional in Naghaay is ’unaan ‘distantly across’. The lexical

interpretation of this term is riverine, but Mr. Pete uses it in a nonriverine situation

(‘across the grass’) when the boy finally finds the frog behind a log in (5.7):

(5.7) Use of ’unaan ‘distantly across’ in Naghaay 54 MP; ’Unaan tl’ogh ta naghaay c’a ’uka nasitelyaesi,

DIST.across grass in frog foc 3pl.look.for.it.rel

kuts’e’ ni c’ayi yaa . to.them 3s.jump

‘The frog they are looking for is jumping across to them on the grass.’


The use of directionals in the two genres differs in both frequency and

literalness. This is attributable to the deictic nature of the directional system. In Saen

tah xay tah and Ta’stede dze’, Mr. Tansy and Mr. Sanford are taking their listeners

on a verbal tour through the Alaskan countryside. Because they are describing

physical geographic locations in relation to a real river, these speakers use

directionals frequently and assign them a literal interpretation.

In Naghaay, however, Mr. Pete uses directionals infrequently, and they are either

interpreted idiomatically or they are limited to the nonriverine uses of ‘up vertically’

and ‘across’. Although there is a river in the fictitious landscape of Frog Where Are

You?, readers have no real awareness of its spatial relationship to the boy’s trek

through the forest until the end of the book. For Mr. Pete to describe the boy’s

direction of travel in terms of the flow of the river would make little sense without a

mental image of its location. Even in the last few pages where Mr. Pete could have

129

used riverine directionals literally – specifically the tumble into the water and the

boy’s subsequent climb onto dry land – he chooses not to do so.

We can already see from the use of directionals that Naghaay lacks some of the

vivid elaboration of direction and location found in the travel narratives. The next

section briefly discuss toponymy, another linguistic domain where travel narratives

are more richly elaborated for space and location than Naghaay. As is discussed,

works of oral literature are often a-geographic in Ahtna.

5.2.3 Toponymy

For their length, the travel narratives contain an impressive number of locations

referenced by their Ahtna toponyms. Mr. Sanford gives twenty-five tokens of place

names over 218 IUs in Ta’stede dze’, and Mr. Tansy gives thirty-one in Saen tah

xay tah over 100 IUs. Not surprisingly, Mr. Pete, when describing the fictional

scenery of Naghaay, gives none.

The reasons for the unequal distribution of toponymy between stories about the

Alaskan countryside and the narration of a children’s book may be self-evident, but

nonetheless the use of place names is vital to creating a sense of space and location

in the travel narratives. While the two genres examined here do not readily lend

themselves to meaningful comparisons of toponym distribution, below I highlight a

few points about Ahtna toponymy to draw attention to its systematicity. Kari has

written extensively on the cognitive and linguistic traits of Ahtna toponymy and its

role in Ahtna geographic knowledge (e.g., Kari 2008, to appear), and the reader is

referred to those sources for more information.

130

Like other aspects of geographic knowledge, mastery of place names occupies a

privileged position in Ahtna culture and identity. Kari stresses that speakers’

attitudes toward the names are consistently cautious and conservative. During his

years spent documenting Ahtna toponyms, Kari found that speakers would prefer to

leave a feature unnamed rather than guess about a name they were unsure about.

Traditionally place names were taught with extreme care and memorized in the

sequence in which one would come across them when traveling along a river or trail.

Naming is also extremely conservative: new names are rarely coined, never

borrowed from non-Athabascan languages, and very frequently shared across

language boundaries with neighboring Athabascan groups (Kari 2008).

The names are structurally systematic and follow a limited number of

conventions. Nearly a third of the corpus is nominalized verbs, and nearly two-thirds

are binomial or trinomial constructions consisting of a specific noun and one or more

generic nouns (e.g., lake, hill, river), as in Kaggos Bene’ ‘swan lake’. Similar names

tend to cluster geographically, such that features in the environs of another more

prominent feature, for example a hill or a lake, will show a recursive naming pattern

based on the name of the prominent feature. Kari (2008) provides an example of

clustering in a set of names in the Reindeer Hills area near Denali National Park.

Yidateni ‘jaw trail’ is the name given to the visually prominent West Reindeer Hill.

Also in the region are an array of features physiographically related to, and taking

their names from, Yidateni: Yidateni Dyii ‘canyon of jaw trail’, Yidateni Dyii

Dghilaaya’ ‘mountain of canyon of jaw trail’, Yidateni Caek’e ‘mouth of jaw trail’,

Yidateni Caek’e Tes ‘hill at mouth of jaw trail’, Yidateni Tl’aa ‘headwaters of jaw

131

trail’, Yidateni Tl’aa Bene’ ‘lake at headwaters of jaw trail’ (2008:27). Mr. Sanford’s

narrative displays some of this toponymic clustering when he talks about locations

near his birthplace at the mouth of the Sanford River. In Ta’stede dze’ he names

Ts’itae Na’ ‘river that flows straight’, Ts’itae Na’ Ngge’ ‘uplands of river that

flows straight’, Ts’itae Caegge ‘mouth of river that flows straight’, and Ts’itae

Tl’aa ‘headwaters of river that flows straight’.

In Ta’stede dze’ and Saen tah xay tah, place names function to orient the listener

to the appropriate geographic region, and the directionals and adverbial prefixes

create a network of paths of motion between them. For speakers and listeners, travel

narratives index a shared knowledge of Ahtna territory, but if listeners are not

personally familiar with a location, the systematicity of Ahtna toponymy allows

them to imagine it. Even if one has never seen the river known as Ts’itae Na’, one

understands immediately that Ts’itae Na’ Ngge’ is the name of its uplands, and that

Ts’itae Caegge is the name of its mouth. The same is often not true of English place

naming conventions. One cannot imagine the physiographic relationship between

Yidateni and Yidateni Na’ based on their English names alone (West Reindeer Hill

and Jack River, respectively).

The absence of place names in Naghaay, on the other hand, is typical of works of

fiction, known as yenida’a, in Ahtna oral literature. Kari observes:

It is quite noticeable that Ahtna yenida’a myths with human-animal interaction are ageographic and always lack place names or any local geographic references. For example, the collection of yenida’a stories by Jake Tansy (1982) contain[s] no place names and can be

132

considered as pure fiction. On the other hand, the presence of place names in narratives appears to be the mark of Ahtna non-fiction. The clan-origin stories, the pre-contact incidents … when two groups of Russians are killed, as well as much earlier regional war incidents (Kari 1986), are non-fiction, prehistoric events that take place at specific places. (Kari 2008:28; emphasis original)

5

We have seen that the speakers in Ta’stede dze’ and Saen tah xay tah make full

use of all three systems described above, while the speaker in Naghaay fully exploits

just the adverbial prefixes, makes only limited use of directionals, and does not need

to use the toponymic system. Naghaay conforms to the a-geographic landscape we

expect from Ahtna fiction, but the source of the difference in spatial elaboration

between it and the travel narratives goes beyond a simple dichotomy between fiction

and non-fiction. Travel narrators and frog-story narrators also have different

narrative tasks, which is reflected in their relative attention to figure and ground; that

is, to the animate referents in the stories and the landscape across which they travel.

The next section discusses how the importance the speakers place on figure and

ground is manifested in differing discourse strategies.

5.3 Attention to Narrative Tasks

Mr. Sanford’s and Mr. Tansy’s richly developed sense of place is consistent with

the sociocultural function of telling travel narratives, which is to index a speaker’s

5 Kari notes that yenida’a do contain directionals even though they are lacking

place names: “The full nine-point system is used, even when the landscape is left to the imagination” (p.c.).

133

intimacy with the land and, by extension, the entitlement of the speaker’s social

group to the resources found there. Speakers pay a great deal of attention to

constructing a sophisticated ground in their stories, and then they move figures

across that ground in a predetermined sequence. The figures are not particularly

differentiated (generally limited to first person plural), but their sequential progress

along the described routes and through the timeline of the story is essential to the

task of telling a travel narrative.

Mr. Pete, on the other hand, is not required to fully elaborate the ground in order

to tell Naghaay. His focus is clearly on the figures. The cast of characters here is

varied and unusual – the boy and the dog are joined by highly agentive wild animals

– and their interactions with one another are the focus of the story. Details about

their path through the forest are unimportant.

The three speakers employ different strategies that reveal what each considers

crucial to the task of telling his story. For Mr. Sanford, progression through the

physical landscape is important, which is reflected in his use of the deictic

postpositional phrase yihwts’en ‘from there’ as a discourse connector. Mr. Tansy is

attuned to the temporal progression of his story, signaled by his use of xona ‘then’ to

connect episodes in his story. Finally, Mr. Pete is most concerned with tracking

individual referents in Naghaay, which he accomplishes via the use of relative

clauses.

134

5.3.1 Discourse Use of the Postpositional Phrase yihwts en

Observe the excerpt in (5.8), in which Mr. Sanford repeatedly uses the

postpositional phrase yihwts’en ‘from there’ (glossed yi-hw-ts’en ‘there-area-from’).

(5.8) Use of yihwits’en ‘from there’ in Ta’stede dze’ 30 AS; Duu yihwts’en, 31 xona ’unggat,

(0.4) 32 Natii Caegge, 33 yedu’ xona,

(1.1) 34 yetdu’ xona nits’ede .

‘From there, then over to ‘Natii Mouth’, then, we would stop there.’

(0.6) 35 Yihwts’en xona Natii Na’ Ngge’,

(2.3) 36 xona ’utgge yii, 37 xungge’ de kude diye.

‘From there, then in ‘Natii River Uplands’, then above there and the uplands are a short distance.’

38 About, (3.1)

39 nduugh miles kulaen. ‘How many miles is it?’

40 Seven, 41 eight miles, 42 I guess.

(1.7) 43 Yet su xona,

(0.3) 44 debae ka ’stede .

‘There we went for sheep.’

(1.3) 45 Debae ts’eghaax.

(0.7) 46 Gha yak’a.

‘We would kill sheep right there.’

135

(1.7) 47 Yii kaen’, 48 taade yet ’sneye .

‘We stayed there three days (living) on it.’

(2.0) 49 Du’ yihwts’en,

(0.9) 50 ts’inats’ede dze’ ’ungge.

‘From there, them we would start out again to uplands.’

(0.9) 51 ’Utggu daagha ngge’,

(1.8) 52 ngga Ts’itae Tl'aa ts’e’,

‘Up above the treeline upland to ‘Headwaters of River That Flows Straight’,’

(1.9) 53 yihwts’en ’unggat,

(1.7) 54 Tsaani ’Ae Na’,

(0.7) 55 yet kets’ede .

‘from there on upland we reached ‘Bear Trap Creek’.’

(1.4) 56 Yet kanaa,

(1.0) 57 debae una’ c’ilaen, 58 you know.

‘Across from there, there are sheep on that creek, you know.’

(0.9) 59 Yet cu debae ka u’stede .

‘There we would hunt again for sheep.’

(0.6) 60 Debae ts’eghaax, 61 you know.

‘We would kill sheep, you know.’

(1.0) 63 Ye naxae ts’eldeli kae,

(0.8) 64 taade nk’e ye ts’eneye .

‘With what we were packing back, we would camp there three days.’

136

(1.4) 65 Duu yihwts’en xona,

(0.4) 66 Natii Na’,

(0.8) 67 Ts’itae Na’, 68 kanats’ede .

‘From there we would go back to ‘Natii River’ and to ‘River That Flows Straight’.’


‘How We Went Hunting Out in the Country’, 00:01:12.440-00:02:17.870. Kari 2010:93-95))

The postpositional phrase yihwts’en ‘from there’ here has a discourse function. It

is used to mark clauses as belonging to the main storyline, which is a listing of the

places on the hunting route Mr. Sanford and his cohort followed. As he names

individual locations, he often digresses to give background information. For

example, in lines 38-48, he first contemplates the distance to the uplands from the

location he has just named, and then mentions that his group would kill sheep and

camp there for three days. He then resumes the main storyline of the journey and

introduces the next two locations, each with yihwts’en, in lines 49-55. He again

provides background information about site usage in lines 56-64, and then continues

along the path to the next location, which is again introduced with yihwts’en in line

65. Each of the twelve occurrences of yihwts’en is used in this way: the discourse

use of this postpositional phrase “gets the characters moving” from place to place,

after Mr. Sanford has departed from the main events of the story to talk a bit about

each location. This discourse use of a spatially-oriented postposition with a deictic

137

demonstrative pronoun as its object highlights the spatial nature of the main

storyline, and allows Mr. Sanford to link episodes of the story together against the

backdrop of the natural landscape.

5.3.2 Discourse Use of the Adverb xona

While Mr. Sanford highlights the spatial ordering of episodes in Ta’stede dze’,

in Saen tah xay tah Mr. Tansy chooses instead to highlight temporal ordering.6 He

does this by linking episodes in his narrative with a discourse use of the sequentially-

oriented adverb xona ‘then.’ This word is clearly related to time; when it is not being

used in a episode-tying function, its lexical meaning is ‘now’. Observe the use of

xona in (5.9).

6 It is not that Mr. Sanford ignores temporal progression; to the contrary, he uses

xona ‘then’ frequently as well. In contrast, however, Mr. Tansy exclusively uses xona. Furthermore, yihwts’en ‘from there’ in Ta’stede dze’ and xona ‘then’ in Saen tah xay tah pattern together in terms of their prosody, suggesting a commonality of function. They tend to occur in intonation unit-initial position, whereas xona in Ta’stede dze’ tends to occur in the middle of intonation units. Prosodic indications of the discourse use, as opposed to the lexical use, of these items warrants further exploration.

138

(5.9) Use of xona ‘then’ in Saen tah xay tah 01 JT; Xona,

(1.0) 02 first, 03 nen’ ta’stghide de’ c’a saen ta,

(0.9) 04 c’a Bes Ggeze Na’, 05 Saas Nelbaay Na’, 06 hwcets’ede .

‘When we first went out in the country during the summer we would ascend ‘Worn Bank Stream’ or “Sand That is Grey Stream’.’

[26 lines about eight locations before reaching ‘Nts’ezi Stream’] 32 Nts’ezi Na’, 33 cu yet cu tcenyii kughi ’aen’, 34 I mean dahtsaa, 35 dahtsaa, 36 hwghi ’a’.

‘At ‘Nts'ezi Stream’ was an underground cache, I mean they had a raised cache.’

37 Teye k’a ’udii, 38 c’etsen’, 39 nkghi ggaasi dahtsaa t’anahghilaes.

‘The meat they had put (there) was enclosed in the pole cache.’

40 Xona ye u Nts’ezi Na’, 41 ye kae na'sdelgges dze’,

‘Then we would come back with that (meat) on ‘Nts'ezi Stream’ and,’

42 dets’en, 43 dets’en, 44 Nts’ezi Na’ ba’aa, 45 dghilaay ghakudaan de kanats’ede n’e , 46 Bes Ggeze Na’.

‘outside of ‘Nts’ezi Stream’ we went back up where a tunnel extends through the mountain, and after ‘Nts'ezi Stream’ we would ascend back up through a canyon in the mountains, and at 'Worn Bank Stream’.’

[eight lines about area around ‘Worn Bank Stream’ and ‘Sand That is Grey Stream’]

139

55 Saas Nelbaay Na’ ngge’, 56 cu ye xona ba’aa, 57 u- uyinanestaani Na’, 58 su hwdedaa’ kanats’ede .

‘Upland of ‘Sand That is Grey Stream’ then again out there we would ascend the downstream area of ‘Stream of the One Protruding Into the Glacier’.’

(2.0) 59 uyinanest'aani Na’, 60 yanaasts'en ’uk’atl’adaak’e cu, 61 Ts’es Ce’e de gaa hwnax, 62 gaani, 63 dighi caax xu dez’aan.

‘On the other side of ‘Stream of the One Protruding Into the Glacier’ is also a bluff ‘Big Rock’ that is as large as this [Jake’s] house.’

(0.5) 64 Ye su xona ’udii,

7

(0.7) 65 hw’e hnats’at’iix,

‘We always used to play there,’

66 hwghak’aay, 67 hw’e u’steltset cu @’snakaey @ts’ghile’ @de @yet.

‘we would run around on the ridge (of the rock) when we were kids.’

(1.1) 68 Yak’a k’adii c’edez’aan.

‘It is still sitting there.’

(1.0) 69 Xona yet u’ ye c’a ye u uyinanest'aani Na’,

(1.4) 70 tsen, 71 tsen tene kana’sghide .

‘Then there at ‘Stream of the One Protruding Into the Glacier’, we go back up to the lowland trail.’


00:00:05.580-00:02:08.170. Kari 2010:60-63))

7 This instance of xona is not an example of its episode-tying function.

140

Like Mr. Sanford’s use of yihwts’en ‘from there’, Mr. Tansy’s use of xona ‘then’

marks transitions between the main storyline of the sequences of arrivals at different

locations on the one hand, and digressions about site use and personal memories on

the other. The story starts with xona (perhaps best translated here at ‘first’), then in

lines 1-31 Mr. Tansy names ten locations. He digresses in lines 32-39 to talk about a

meat cache. He resumes the storyline with xona in line 40, where the next event is

the return via ‘Nts’ezi Stream’ with meat from the cache. The arrival at ‘Stream of

the One Protruding Into the Glacier’ is marked with xona in lines 55-58, followed by

ten lines of personal recollections. Again, the journey is resumed in line 69 with

xona.

5.3.3 Tracking Referents with Relative Clauses

The discourse strategies of the travel narrators underline the importance they

place on ground, as opposed to figure. The digressions consistently provide

background information about locations, rather than about the people traveling

through them. In fact, the travel narrators give very little information about the

characters in these stories, and almost exclusively refer to them with subject prefixes

only. Mr. Pete, on the other hand, is far more concerned with figure than with ground

in his frog-story narrative. He attends carefully to the task of tracking characters as

they interact with each other in the minimally defined landscape of Naghaay. He

does this most notably by using relative clauses to refer to and delimit referents that

have already been introduced. See (5.10).

141

(5.10) Use of relative clauses in Naghaay 19 MP; Ngga t’ox naggic’e ’.

upland nest 3s.hang.down

‘Up (over there) the nest is hanging.’

20 Cii c’e’an ugha niyaa. boy den to.it 3s.come

‘The boy comes to a den.’

21 ic’ae ngga t’ox naggic’e ’i gha itsae. dog upland nest 3s.hang.down.rel at 3s.bark

'The dog barks at the nest that is hanging (there).'

22 Dligi kaniyaa, squirrel 3s.go.up.and.out

‘A squirrel comes up,’

23 ic’ae ngga t’ox naggic’e ’i gha itsae. dog upland nest 3s.hang.down.rel at 3s.bark

‘the dog barks at the nest that is hanging (there).’

24 Dligi c’a kaghiyaani, squirrel foc 3s.go.up.rel

25 ic’ae hna ’aen’. dog 3s.see

‘The squirrel who came out is looking at the dog.’

[…] 47 Tadedze’ ce’e yii kena ’aen,

driftwood big in 3pl.see

‘They see a big piece of driftwood,’

48 ic’ae utse’ k’e dayizdaa, dog his.head on 3s.sit

‘he dog is sitting on his head,’

49 tadedze’ gha nihnidaetl’, driftwood to 3.pl.arrive

'they arrive at the driftwood,'

50 tadedze’ nahditaani ye kiigha delts’ii. driftwood they.found.rel there by 3p.sit

‘they are sitting by the driftwood they found.’

[…] 52 Tsets nahditaani k’e dahdelts’ii.

wood they.found.rel on 3p.sit

‘They sit down on the wood they found.’

[…] 54 ’Unaan tl’ogh ta naghaay c’a ’uka nasitelyaesi,

across grass in frog foc 3pl.look.for.it.rel

142

kuts’e’ ni c’ayi yaa . to.them 3s.jump

‘The frog they are looking for is jumping across to them on the grass.’

((Markle Pete, Naghaay Ndaane Zidaa ‘Frog Where Are You’, oai.paradisec.org.au:ALB01-AHTNB001-060.tiff, ALB01-AHTNB001-061.tiff,

ALB01-AHTNB001-065.tiff, ALB01-AHTNB001-066.tiff))

The relative clauses in lines 21 and 23 refer to the bees’ nest, which had been

introduced in line 19. The relative clause in 24 refers to the squirrel introduced in

line 22. The relative clauses in 50 and 52 refer to the driftwood that had been

introduced in line 47, and the relative clause in 54 refers back to the frog, which had

been introduced at the beginning of the story. Note that in terms of cognitive

activation states of referents (e.g., Chafe 1994), such careful tracking may not be

strictly necessary in all cases. For instance, the squirrel first appears in line 22, and

only one line intercedes between its appearance and the use of a relative clause to

refer to it. There is no chance here for confusion with another squirrel, but Mr. Pete

packages it carefully just the same. Similarly, the driftwood is introduced in line 47,

referred to again in line 49, and then delimited with a relative clause in line 50.

Mr. Pete’s approach to the tasks of narrating Naghaay is different from that of

the travel narrators. At no point does he depart from the storyline, nor does he use

yihwts’en, xona, or any other such marker to contrast storyline clauses with

digression clauses. But where Naghaay is lacking in discourse markers and

digressions about locations, Ta’stede dze’ and Saen tah xay tah are plainly lacking

143

in relative clauses and elaborate tracking of characters.8 Note that the travel narrators

were not asked specifically to avoid stories “about people”—rather, this genre is

inherently about landscape, travel, and events over detail tracking of animate

referents. The speakers here choose grammatical mechanisms for elaborating figure

or ground that are consistent with the tasks they deem necessary for storytelling.

5.4 Conclusion

Typologists have used frog story narration to compare how languages express the

notion of direction in motion events, and to make predictions about how a language

is likely to behave based on those comparisons. As we have seen, Ahtna frog story

narration does give us a glimpse into the resources of the language for expressing the

notion of direction: Mr. Pete makes extensive use of the semantically general

direction-describing adverbial prefixes. We have also seen that there is much about

the grammar of direction that Naghaay does not reveal. Had we relied solely on frog

stories to tell us about Ahtna encoding of direction and direction in motion events,

the descriptive richness and frequent use of the use of directionals and toponymy in

the travel narratives would have remained hidden. Indeed, omitting either of these

from a discussion of motion events would result in a poor description of Ahtna

grammar.

Of course, the goal of typological frog story research is not to develop

comprehensive descriptions of the grammar of direction and location for any single

8 Except for those that have lexicalized into toponyms, relative clauses occur

only once in each travel narrative.

144

language, but to provide semantically unified content for cross-linguistic

comparisons. Frog stories are attractive because they provide samples of connected

speech, but because they are not the vivid, lived experiences relayed in the travel

narratives, they are less likely to reveal what is most natural in discourse. Frog

stories are told in a highly contrived setting. The narration of a children’s book is not

an indigenous genre of Ahtna discourse in the way that the telling of travel narratives

is, a point that was driven home by a female Ahtna consultant who refused to

participate because “Ahtna people don’t keep frogs as pets.” We should be careful to

include in any description examples from discourse that is more typical of the

language community in which the grammatical structures we are investigating arose.

In sum, there is nothing linguistic that prevents travel narratives from being focused

on people rather than locations and events, but the traditional genre of travel

narration comes with an ideology that influences how speakers use the linguistic

resources available to them.

Finally, we also need to consider what speakers are actually attending to. During

narration of a frog story, it is likely that unless the speaker is savvy enough to

understand that a researcher is investigating the particulars of how the language

segments direction and manner in motion events, he or she will be attuned to tasks

other than providing a good sample for such research. It is far more likely that when

asked to narrate the storybook, a speaker will try to do just that: to convey the events

in the book in the order in which they happen with attention to whatever factors

seem most important. For Mr. Pete, creating richly imagined characters and keeping

track of them through the series of unusual events is important. For Mr. Sanford and

145

Mr. Tansy, creating highly elaborated landscapes and providing background

information and personal memories is important. If a frog story-narrator does not

consider elaborate descriptions of direction to be essential to the storyline, he or she

may leave them out in favor of other concerns. Thus we need to cast a wide net when

making typological observations and take into account data from a range of sources

(e.g., Applebaum & Berez 2009). Ultimately it is not essential for frog story

narrators to create a fully fleshed out sense of landscape, which can hide aspects of

the grammar from us.

The next two chapters examine the morphology of the directionals, as they are

used in spontaneous speech, more closely.

146

Chapter 6

Lexicalization of the Directional Prefixes:

Evidence from Discourse

6.1 Introduction

The previous chapter discussed the prevalence of path and location description in

Ahtna travel narratives. This chapter and the next examine the morphology of the

directional words, with particular attention to how they are used by speakers in

spontaneous discourse. The goal of these chapters is to begin to address some

preconceptions about the nature of polysynthesis using the Ahtna directionals as a

case study. To illustrate, consider the following quote from Duponceau (1819), the

coiner of the term polysynthesis:

The manner in which words are compounded in that particular mode of speech [i.e., polysynthesis], the great number and variety of ideas which it has the power of expressing in a single word; particularly by means of the verbs; all these stamp its character for abundance, strength, and comprehensiveness of

expression, in such a manner that those accidents must be considered as included in the general descriptive term polysynthetic (Duponceau 1819, emphasis added)

In other words, polysynthetic languages, which are those that have a high

morpheme-to-word ration, are considered by Duponceau to be by definition capable

of expressing an extraordinary amount of information in a single word and

147

characterized by a comprehensiveness of expression. While it is true that

polysynthetic languages express in morphemes information that analytic languages

express in words (e.g., argument structure, mood, aspect, semantic roles), the

assumption that polysynthetic words are remarkable in their expressivity is just that:

an assumption. In metalinguistic situations with speakers of polysynthetic languages,

for example elicitation, it may seem to an English-biased researcher that the many

permutations of polysynthetic words that are judged acceptable in that context are

also accessible to speakers in spontaneous language use, but that claim is something

that can only be borne out by an examination of discourse.

In this chapter and the next, I examine the use of the polysynthetic directionals in

spontaneous Ahtna discourse. Recall from Chapter 2 that the directionals have a

tripartite structure: a stem expressing orientation (a system that is largely, but not

completely, riverine), an optional prefix expressing relative distance or concepts like

‘straight’, ‘adjacent to’, etc., and an optional suffix that expresses either a punctual

vs. areal distinction or an allative vs. ablative distinction (see Kari 1985, 1990, 2008,

2010; Leer 1989; Moore & Tlen 2007) This is shown in Table 6.1, repeated from

Table 2.1 in Chapter 2.

148


(from Kari 1990)










nse’ ‘ahead’

In what follows, I show that contrary to assumptions about polysynthesis, in

discourse speakers do not make use of the full “comprehensiveness of expression”

that is morphologically available to them in the directionals. The deictic prefixes are

in fact less expressive than their historical sources suggest. I show that the prefixes

are undergoing semantic bleaching and lexicalization, a natural process of language

change. I present evidence from geographic information systems technology that

shows that the deictic prefixes no longer map to a three-way distinction in relative

distance. In Chapter 7 I discuss another process of language change that is occurring

in the directional suffixes; namely, these are being extended to discourse functions,

building structure into the narrative and interestingly recapturing some of the

semantic distinctions that are being lost in the prefixes.

Section 6.2 below presents the data and methodology for this and the following

chapter, and describes how the complexity of directional use in Ahtna discourse

leads to difficulty in interpreting the meaning of each individual directional

morpheme. Section 6.3 presents the literature (relying heavily on Leer (1989)) on the

149

historical sources of the directional stems and affixes. Section 6.4 frames the debate

over compositionality and productivity in the Ahtna directionals, and presents the

possibility that general forces of language change, namely lexicalization, are causing

the directionals to become less analyzable than might be expected. Section 6.5

presents data on the uses of the deictic directional prefixes, including evidence from

geographic information systems technology that shows the range of distances

described with the deictic prefixes. Section 6.6 contains a discussion of the semantic

bleaching and lexicalization found in the directionals as they are used in discourse,

and segues to the next chapter, which discusses the use of the directional suffixes to

structure travel narrative discourse and provide distinctions between distances of

different lengths.

6.2 Directional Morphology in Discourse

As an example of the complexity of directional morphology in discourse,

consider (6.1), which is an excerpt from a travel narrative.

(6.1) Use of directionals in discourse 46 Du’ yihwts’en,

there.AREA.from

(0.2) 47 ts’inats’ede dze’ ’unggeh,

1PL.SUB.pl.go.IMPERF.from.ITER thus DIST.upland.ALL

‘From there, then we would start out to uplands.’

(0.8) 48 ’utggu daagha ngge’,

DIST.up.AREA above.timberline upland

(1.7) 49 ngga Ts’itae Tl’aa hwts’e’,

upland.PUNC river.flows.straight.headwaters AREA.to

‘Up above the treeline upland to ‘headwaters of river that flows straight’ (Sanford River headwaters),’

150


there.AREA.from DIST.upland.PUNC


bear.trap.river


there 1PL.SUB.pl.go.IMPERF.against.a.place

‘from there on upland we reached ‘bear trap creek’.’

((Adam Sanford, C’uka Ts’ul’aen’i gha Nen’ Ta’stede dze’ ‘How We Went Hunting Out in the Country’,

00:01:45.390-00:01:59.290. Kari 2010:93))

This example of seven IUs contains five directionals totaling twelve morphemes.

The directionals and glosses of the morphemes they are composed of are shown in

Table 6.2.

Table 6.2: Directional Morphology Found in (6.1)

IU Directional Prefix Stem Suffix

47 ’unggeh DIST upland ALL

48 ’utggu DIST up (vertically) AREA

48 ngge’ - upland -

49 ngga - upland PUNC

50 ’unggat DIST upland PUNC

The region covered in the excerpt in (6.1) is shown in Map 6.1. Mr. Sanford

references two locations by their Ahtna toponyms Ts’itae Tl’aa (a) and Tsaani ’Ae

Na’ (b), and a third by the deictic demonstrative pronominal prefix yi- ‘there’ in IU

46 that is coreferential with a previously named location, Natii Na’ (c).

151

Map 6.1: Area Described in (6.1)

Questions arise about how to interpret this excerpt. We already know that Mr.

Sanford’s narrative follows the course of the Sanford River, but how does the

listener know in what order the named locations are encountered, and why does Mr.

Sanford choose to describe the route with this particular collection of directional

prefixes, stems, and suffixes? Why are his destinations referred to sometimes with

the allative directional suffix, sometimes with the area suffix, and sometimes with

the punctual suffix? Is his alternation between the use of the distal prefix and the

omission of any prefix meaningful? If the general trajectory of this portion of Mr.

Sanford’s narrative is from northwest to southeast, why does he mention Ts’itae

152

Tl’aa before he mentions Tsaani ’Ae Na’? These questions will be addressed in this

chapter and the next.

6.2.1 Data and Methodology

This chapter and the next examine the morphology of Ahtna directionals as they

are found embedded in discourse in which the speaker’s task is to describe routes of

travel through known portions of Ahtna territory. The data presented here come from

three travel narratives found in Kari 2010. The first is from Adam Sanford’s

narrative C’uka ts’ulaen’i gha nen’ ta’stede dze’ ‘how we went hunting out in the

country’ (Kari 2010:91-128), described in Chapter 5. The first eight minutes and

twenty seconds of this recording are included here, which correspond to 218 IUs.

The second narrative is an Upper Ahtna dialect narrative by Katie John, known as

Natae Nenn’ ‘Batzulnetas country’ (Kari 2010:74-89). This recording is

approximately 12:40 in length and corresponds to 339 IUs. The third narrative is

Jake Tansy’s Saen tah xay tah c’a u’sghide ‘we used to travel around in summer

and winter’ (Kari 2010: 59-69), also described in Chapter 5. The first three minutes

and five seconds are considered here, equivalent to 100 IUs.

Because we already know the communicative goal of Ahtna travel narration is to

accurately describe routes of travel through real locations in Ahtna territory (cf.

Chapter 5), a study of how speakers select morphology requires consideration of the

physical spatial relationships among those locations. To this end, this chapter and the

next rely heavily upon geographic information systems (GIS) technology and Kari’s

(2008) database of more than 2200 georeferenced Ahtna toponyms. The database is

153

the product of over three decades of Kari’s work with Ahtna speakers to collect,

map, and verify the geographic coordinates and names of locations both inside and

outside of Ahtna territory, and it represents the most comprehensive geographic

database for any language in Alaska. Kari generously provided the digital version of

the database to me, which I uploaded to Google Earth Pro software. Using the

mapped toponyms and the travel narrative texts as a guide, I then mapped the routes

described in the three narratives in the Google Earth Pro interface. This mapping

allows us to visualize how the deictic directional prefixes are being used, and

whether or not they correspond in discourse to a three way near vs. intermediate vs.

far distinction. Is it crucial to note, however, that the Kari database has locations

mapped only as points even when in reality it would have been better to map them as

polygons, lines, and fuzzy-bounded areas. Therefore all distances given in this

chapter and the next are understood to be approximate only. An overview of the

regions covered by all three narratives is shown in Map 6.2, and individual routes are

shown in Map 6.3, Map 6.4, and Map 6.5.

154

Map 6.2: Overview of Travel Routes Described in the Three Narratives:

Mr. Sanford (blue), Mrs. John (yellow), and Mr. Tansy (pink)

155

Map 6.3: Route Described in Mr. Sanford’s Narrative

Map 6.4: Route Described in Mrs. John’s Narrative

156

Map 6.5: Route Described in Mr. Tansy’s Narrative

Because this is a study of directional morphology, a more comprehensive

treatment of the historical development of the directional suffixes, prefixes, and

stems is required here; these are presented in Section 6.3 below.

6.3 Historical Sources of the Directional Morphology

Leer 1989 is the most thorough treatment of the historical development of

directional systems in the Na-Dene family. He postulates that the directional stems

were historically free, unpossessed nouns, to the right and left of which additional

material could be added to form adverbs or adverbial phrases. Reconstructions from

Leer (1989) of the Proto-Athabascan directional prefixes are presented in Section

6.3.1, reconstructions of the Pre-Proto-Athabascan directional stems are presented in

157

Section 6.3.2, and reconstructions of the Proto-Athabascan directional prefixes are

presented in Section 6.3.3.

6.3.1 Historical Sources of the Directional Prefixes

According to Leer, the historical source of the deictic prefixes found on modern

directionals are “historically demonstrative stems that have lost independent status

and become part of the following material” (Leer 1989:593). Leer’s tentative

reconstructions of these demonstratives and their Ahtna reflexes as directional

prefixes are given in Table 6.3 below.

Table 6.3: Proto-Athabascan demonstrative prefixes

and their Ahtna reflexes as directional prefixes

PA forms (from Leer 1989)

Ahtna reflexes (from Kari 1990)

1

*da·- ‘nearby’ da- PROX

*na·- ‘not distant’ na- MED

*yu·- or * u·- ‘distant’ ’u- DIST

*c’i- ‘straight (in the direction indicated)’

2

ts’i- ‘straight, directly’

As will be discussed, the three deictic prefixes (i.e., ’u- ‘DIST’, na- ‘MED’, and

da- ‘PROX’) no longer show a clear three-way distinction in discourse, despite the

semantics of their historical sources.

1 This is not the complete list of prefixes that are found on directionals, but just

the ones for which Leer (1989) provides reconstructions. Kari’s complete (1990) list can be found in Table 2.1 in Chapter 2.

2 Leer notes that the prefix denoting ‘straight’ is not a demonstrative, but that it

can be grouped with the other prefixes (presumably on distributional grounds).

158

6.3.2 Historical Sources of the Directional Stems

Leer’s reconstructions of the Pre-Proto-Athabascan roots from which modern

Ahtna directional stems arose are given in Table 6.4 below.

Table 6.4: Pre-Proto-Athabascan Directional Stems

and Their Ahtna Reflexes

PPA forms (from Leer 1989)


**ni - ‘upriver, back behind, to the rear’

nae’ ‘upriver, behind’

**da - ‘downriver’ daa’ ‘downriver’

**n - ‘from water’ ngge’ ‘from water, upland’

**c n - ‘down to shore’ tsen ‘toward water, lowland’

** a·n - ‘across’ naan ‘across’

— tgge’ ‘up vertically’

**d - ‘down below’ igge’, yax ‘down vertically’

** n - ‘away’ ’an ‘away, off’

**n sd- ‘ahead’ nse’ ‘ahead’

Leer sums up his discussion of the development of the directionals with a note

about their phrasal status:

At an early formative stage of Athabaskan, as in Tlingit, I postulate that the directional stem was lexically a noun that could be preceded by a demonstrative and to which postpositions [and adverbializers] were added to form adverbial phrases. Compare the following sentences, both translating roughly [as] ‘he walked way up there’:

Tlingit: yú dá· -i kè uwagúd yonder upland-LOC up he.walked

Koyukon: yu-d g- xo-lyo yonder-above-ALL up-he.walked

In Tlingit, yú dá· ‘yonder upland’ is a NP just like yú

àn ‘yonder town’; the addition of the postposition –i to this NP […] results in an adverbial phrase. This is

159

exactly what I wish to claim was originally true in Athabaskan. (Leer 1989:594f)

Thus the directional stems were historically roots in Pre-Proto-Athabascan,

which was treated like a noun in that it could take demonstratives and postpositions.

The next section presents the historical source of the directional suffixes.

6.3.3 Historical Sources of the Directional Suffixes

Historically, postpositions or adverbializers denoting the semantic distinctions

presented in Table 2.1 in Chapter 2 (i.e., allative, ablative, punctual, or areal) were

found to the right of the free directional nouns. Adverbializers were likely more

frequent than postpositions, and Leer does not provide reconstructions of historical

directional noun+postposition forms, but only notes that speakers today “sometimes”

add postpositions to directional stems (Leer 1989:591). In the travel narratives

examined here, directionals with postpositions as suffixes do occur but are indeed

rare. Only two such examples occur, both with suffixed ts’en ‘side of’. These are

shown in (6.2) below.

(6.2) Ahtna directionals suffixed with postpositions (as opposed to adverbializers) (a) 09 Kolgiis Na’ datsiists’en de.

K.river PROX.downland.side.of ADVZ.TEMP.SPATIAL

10 Yihwk’e, there.AREA.on

(0.9) 11 ni t’aay kade dze’.

RECIP.front.of 3PL.SUB.go.IMPERF thus

‘They used to meet on the lowland side of Kolgiis Creek […]’


160

(b) 73 ’Unggat,

DIST.upland.PUNC

(0.2) 74 Xoos Ghadl Zdlaa keniide ye

horse.wagon.exist 3PL.SUB.say.IMPERF.WHERE there

danggasts’en ta, PROX.upland.side.of among

(0.6) 75 xona danggehdze dae’

then PROX.upland.ABL thus

tic’akede . 3PL.SUB.pl.go.IMPERF.out.into.the.woods

‘Up at where they call it ‘horse wagons are there’ there then they would go out into the country to the upland side or from the uplands.’


As for the Proto-Athabascan adverbializers, Leer does not state explicitly that

these were ever historically free morphemes, but he does note that in Proto-

Athabascan their phonological erosion was so strong that they became semantically

somewhat opaque. “It is thus often more appropriate to view the original

combination of stem plus postposition as a stem plus unclassified suffix or as an

unanalyzable stem” (Leer 1989:591). Leer’s reconstructions of the Proto-Athabascan

adverbializers and their Ahtna reflexes as directional suffixes are given in Table 6.5

below.

161

Table 6.5: Proto-Athabascan Adverbializing Suffixes

and their Ahtna Reflexes as Directional Suffixes

PA forms (from Leer 1989)


*-x ‘toward’ -e ALL

*- ’ n ‘from’ -dze ABL

*-d ‘location at a specific point’

-t PUNC

*-x or *- ‘location in the area’

-xu AREA

Now that we have seen the historical sources of the modern Ahtna directional

prefixes, stems, and suffixes, the next section turns to a discussion of

compositionality in speakers’ use of directionals in discourse.

6.4 Compositionality in the Directional Morphology

In contrast to Leer’s (1989) quote above about the unanalyzability of the

directional stem+suffix combination, Kari has long advocated for the productivity of

the directional morphology. His earliest writings on the topic present high numbers

of possible directional types. In Kari 1985 he writes, “A directional (or locational)

word in an Alaskan Athapaskan language is made up of three elements: (1) one of

six prefixes […], (2) one of nine directional roots […], and (3) one of four suffixes

[…] When prefixes, roots, and suffixes are combined, 216 directional words occur”

(p. 471-472, emphasis added). Similarly in Kari 1989, he writes “in Ahtna there is a

set of nine directional roots, which can be derived into 216 directional words” (p.

141). These figures reflect his early understanding of the morphemes that are

available in each position, and by 1990 he settles on the list of eight prefixes, nine

162

stems, and four suffixes given in Table 2.1 in Chapter 2. Interestingly, from that list,

the total number of logically possible directional tokens in fact comes to 405. While

more recent writings by Kari (2010, to appear) significantly scale back claims about

the number of possible derived directional words,3 his enthusiasm for the

combinatoriality of the system is not abated:

With the plain [stems] (without prefix or suffix) and the combinations of prefixes, roots, and suffixes there are sixty-nine [sic, see fn. 3 below] possible directional terms. Expert speakers can tap into all of these fine distinctions […] the Ahtna experts certainly

distinguish more than sixty precise meanings with these words” (Kari 2010:130, emphasis added).

Busch (2000) likewise alludes to full combinatoriality of Gwich’in directionals,

stating that the nine directional stems in that language take one of four suffixes and

one of five prefixes. Moore & Tlen (2007) suggest an even larger range of possible

combinations for Southern Tutchone directionals, which also allow prefixing of

personal pronouns.

In the 657 IUs of the three travel narratives examined here, 135 tokens covering

63 types of directional words (i.e., combinations of prefixes, stems, and suffixes) are

found. I am not disputing Kari claims that speakers can perceive differences among

the more than 400 different logically possible combinations of directional

3 The figure given in both Kari 2010 and Kari, to appear, is 69. However, it is

unclear how this figure is reached. The morphemes listed in these publications include 5 prefixes, 9 stems, and 5 suffixes, and directional words can occur as (i) bare stems, (ii) prefix+stem combinations, (iii) stem+suffix combinations, and (iv) prefix+stem+suffix combinations. It is possible that (a) Kari is being conservative

163

morphology; especially savvy speakers may even be able to produce a great number

of forms during elicitation sessions, especially when the task is presented in the form

of a paradigm. However, elicitation exposes a metalinguistic knowledge that is

unlike the production of spontaneous speech, and it is important to understand why

speakers do not produce such a wide variety of forms in discourse, even when

describing regions of great physiographic diversity.

Table 6.6 shows the fourteen types in the data with a token frequency of three or

higher.

Table 6.6: Most Frequent Directional Words in the Travel Narrative Data

in Descending Order of Frequency

Token Frequency

Type(s)

17 ’unggat ‘DIST.upland.PUNC’

11 ngge’ ‘upland’

6 ’utsen ‘DIST.downland’

xungge ‘AREA.upland.ALL’ 5

’ungge ‘DIST.upland.ALL’

4 nanaan ‘MED.across’

nae’ ‘upriver’

’unaat ‘DIST.across.PUNC’

’utsiit ‘DIST.downland.PUNC’

’udaa’ ‘DIST.downriver’

’uyggu ‘DIST.up.AREA’

’unggu ‘DIST.upland.AREA’

datsuughe ‘PROX.downland.AREA’

3

danggu ‘PROX.upland.AREA’

Table 6.7 shows the frequencies of the directional prefixes in the travel narrative

data.

and only counting tokens that actually occur in the published travel narrative corpus,

164

Table 6.7: Frequencies of Directional Prefixes in the Travel Narrative Data

in Descending Order

Frequency Type(s)

55 ’u- ‘DIST’

35 (no prefix)

19 da- ‘PROX’

10 hw- ‘AREA’

na- ‘MED’ 8

ka- ‘adjacent’

Table 6.8 shows the frequencies of the directional stems in the travel narrative

data.

Table 6.8: Frequencies of Directional Stems in the Travel Narrative Data

in Descending Order

(bare stem given for reference, data includes affixed forms)

Frequency Type(s)

47 ngge’ ‘upland’

29 tsen ‘downland’

15 nae’ ‘upriver’

14 naan ‘across’

12 igge’ ‘up’

11 daa’ ‘downriver’

4 ’an ‘away’

2 tgge’ ‘down’

1 nse’ ‘ahead’

Table 6.9 shows the frequencies of the directional suffixes in the travel narrative

data.

or (b) this figure is simply a miscalculation.

165

Table 6.9: Frequencies of Directional Suffixes in the Travel Narrative Data

in Descending Order

Frequency Type(s)

49 (no suffix)

37 -t ‘PUNC’

18 -xu ‘AREA’

16 -e ‘ALL’

13 -dze ‘ABL’

2 -ts’en ‘side of’

Table 6.7 shows that ’u- ‘DIST’ is the most common prefix found in the data,

followed by da- ‘PROX’, hw- ‘AREA’, na- ‘MED’, and ka- ‘adjacent’. Table 6.8 shows

that ngge’ ‘upland’ is the most common directional stem in the data, followed by

tsen ‘downland’ and nae ‘upriver’. Table 6.9 shows that unsuffixed forms are most

common in the data, followed by those suffixed with -t ‘PUNC’, -xu ‘AREA’, -e

‘ALL’, and -dze ‘ABL’.

If indeed the directional morphology is, as Kari suggests, highly combinatorial,

we would expect to find more than fourteen frequent types in the data. Why then

don’t we? The historical sources of modern Ahtna directional morphology are fairly

transparent to researchers who are familiar with the literature on Athabascan

historical linguistics. We must, however, consider that over the course of the

development of modern Ahtna, the forces of language change have been at play, and

that directionals are as susceptible to phonological fusion and semantic bleaching as

166

any other part of the grammar. We must be cautious not to conflate historical

analyzability with synchronic analyzability.4

The linguistic process by which grammatical morphemes fuse to lexical

morphemes and lose their semantic richness goes by a variety of names including

lexicalization, univerbation, and degrammaticalization (e.g. Hopper & Trauggot

2003; lexicalization will be used here). According to Lipka (1990), lexicalization is

the process by which “a complex lexeme once coined tends to become a single

complete unit, a simple lexeme” (p. 95). In lexicalization, grammatical morphemes

become semantically and often phonologically inseparable from their lexical hosts.

Phonological fusion of the Proto-Athabascan demonstratives and the Pre-Proto-

Athabascan adverbializers has already resulted in their becoming affixes to the

directional stem. The low type count (i.e., fourteen types of a frequency of three or

more) in the corpus could then be explained by lexicalization if it could be shown

that semantic bleaching of the affixes is also occurring. This would mean that

speakers are not distinguishing the affixes from one another in ways that the

established semantics of their historical sources suggest. Section 6.5 below shows

that in travel-narration discourse speakers are not using the deictic prefixes in a way

that supports a clear distal vs. medial vs. proximal distinction (’u-, na-, and da-

respectively, see Table 6.1). Rather than reserving the distal prefix for comparatively

long distances, the medial prefix for comparatively intermediate distances, and the

4 In the words of Sally Rice (p.c.) regarding Athabascan polysynthesis in general,

“beware the conceit of analyzability.”

167

proximal prefix for comparatively close distances, even within single route

descriptions by individual speakers, the deictic prefixes are used to describe routes of

significantly varied length. This loss of semantic transparency points toward a

process of lexicalization in the prefixes, which may in part account for the lower

than expected type count in the travel narratives.

6.5 Mapping the Use of Deictic Prefixes in the Travel Narratives

This section provides an overview of how the three narrators use the deictic

directional prefixes to describe legs of their journeys, in particular to describe

differences in distance that the deictic prefixes purportedly distinguish. As a

reminder, the deictic prefixes are ’u- ‘DIST’, na- ‘MED’, and da- ‘PROX’. Given the

route-description purpose of the travel narratives, the assumption here is that deixis

refers to locations mentioned in the narratives (rather than the speaker’s location).

For each of the three prefixes, text examples and maps are given to show the variety

of physiographic situations the directionals describe.

6.5.1 The Distal Prefix u-

As is shown in Table 6.7, the distal prefix ’u- is by far the most frequent prefix in

the data. Individual speakers’ usages of this prefix are discussed in turn below.

6.5.1.1 Mr. Sanford s Use of the Distal Prefix u-

Mr. Sanford uses the distal prefix ’u- 22 times in his narrative, referring to

fourteen separate legs of the journey. The geographic distance to which he refers

with these morphemes ranges from 1.5 miles to 24.3 miles (mean 9.32 miles). Table

168

6.10 shows a summary of the journey legs, by point of departure and point of arrival,

that are described with directionals prefixed with ’u-, as well as their approximate

length in miles. A selection of these legs is illustrated with maps and their

corresponding text below.

Table 6.10: Journey Legs Described by Mr. Sanford Using ’u- ‘DIST’

and Their Length in Miles

Leg Description

Length in miles

(approximate) in ascending

order

Ben Tah downland to shore of Copper River 1.5

Ts’itae Caegge (mouth of Sanford R.) overland to Ba’stade i 2

Natii Caegge (mouth of Natii R.) upstream to Natii Na’ Ngge’ (Natii R. uplands)

2.4

Mouth of Una’ Hwnelk’ezi Na’ (river) to sheep hunting grounds on far side of Copper River.

3.6

Mouth of Una’ Hwnelk’ezi Na’ (river) upstream to its glacier 4.7

Mouth of Tsaani ’Ae Na’ to mouth of Una’ Hwnelk’ezi Na’ 5.1

Up Kaghaa Na’ river from mouth to headwaters 7.1

Mouth of Natii Na’ (river) to Tsaani ’Ae Na’ (river) 7.8

From Ba’stade i downland to Ts’itae Na’ (Sanford R.), then along Ts’itae Na’ to Natii Caegge (mouth of Natii R.)

9.3

Mouth of Una’ Tuu Koley Na’ (river) upstream to Saas Dze 9.5

Mouth of Una’ Hwnelk’ezi Na’(river), along Ts’itae Na’ (Sanford R.), and up Kaghaa Na’ (river) to its headwaters

13

Mouth of Tsidghaazi Na’ (river) to its headwaters at Tsidghaazi Tl’aa

19

From Natii Caegge (mouth of Natii R.) along Ts’itae Na’ (Sanford R.) to its headwaters at Ts’itae Tl’aa.

21.2

From Saas Dze down Una’ Tuu Koley Na’ (river) to its mouth, then up the Copper River to Mr. Sanford’s fish camp near mouth of Kedi eni Na’ (river).

24.3

The shortest leg Mr. Sanford describes with a directional containing ’u- ‘DIST’ is

shown in Map 6.6, with the corresponding text given in (6.3). In this leg, Mr.

169

Sanford is describing a short walk downland from Ben Tah to the shores of the

Copper River; he uses ’utsuughe ‘to distantly downland’ in IU 114 of (6.3).

Map 6.6: Example of ’u- ‘DIST’ for the Route from

Ben Tah to the Copper River (1.5 miles)

(6.3) Mr. Sanford’s use of ’u- ‘DIST’ to describe comparatively short distance 109 Xona dae’ ts’en,

then thus from

(3.1) 110 ben,

lake

(2.8) 111 ben ta-,

lake ABAN

112 Ben Tah keniide. lake.among 3PL.SUB.say.IMPERF.where

170

(0.8) 113 Yet c’a yii (4.0) nits’enaes.

there FOC 3S.NH 1PL.SUB.move.nomadically.IMPERF.TERMIN

‘Then on that side there, a lake, where they call ‘among the lakes’, there we stopped.’

114 Xona ’utsuughe, then DIST.downland.ALL

(2.0) 115 ts’abaeli nae’ nel’aa xu,

tree upriver linear.extends AREA

(0.8) 116 xu,

AREA

(0.5) 117 xu ninats’etnaes.

AREA 1PL.SUB.move.nomadically.IMPERF.TERMIN.ITER

(0.3) 118 Xona deniigi ka.

then moose in.quest.of

‘Then we came to the [distant] lowland area where the treeline extends upstream (direction toward upper Copper River), and we stopped again, then for moose.’


00:04:08.130-00:04:31.170. Kari 2010:96))

An example of Mr. Sanford’s use of ’u- to describe a distance of intermediate

length is shown in Map 6.7 below. This leg begins at Ba’stade i, nearly due west of

the junction of the Copper and Sanford Rivers. The route heads overland to Ts’itae

Na’ (Sanford River), then along Ts’itae Na’ to Natii Caegge (mouth of Natii River).

This leg is approximately 9.3 miles long.

171


Ba’stade i to the Natii Caegge (9.3 miles)

As an example of ’u- ‘DIST’ showing a route of intermediate distance in (6.4)

below, Mr. Sanford uses ’unggat ‘a point distantly upland’ to describe the direction

from Ba’stade i to Natii Caegge.

(6.4) Mr. Sanford’s use of ’u- ‘DIST’ to describe comparatively intermediate distance 28 Duu yihwts’en xona ’unggat,

there.AREA.from then DIST.upland.PUNC

(0.3) 29 Natii Caegge,

N.rivermouth

30 yedu’ xona, then then

172

(1.0) 31 yedu’ xona nits’ede .

then then 1PL.SUB.pl.go.IMPERF.TERMIN

‘From [t]here [i.e., Ba’stade i], then [to a point distantly upland] to ‘natii mouth’, the we would stop there.’


00:01:12.440-00:01:18.370. Kari 2010:93))

The longest journey leg Mr. Sanford describes using’u- starts at a location called

Saas Dze , down a river called Una’ Tuu Koley Na’, and up the Copper River to his

fish camp near Kedi eni Na’ (Map 6.8 below). The route is approximately 24.3 miles

long. Mr. Sanford uses ’uniit ‘a point distantly upriver’ twice (IUs 175 and 177) in

the corresponding excerpt shown in (6.5).


Saas Dze to Fish Camp Near Kedi eni Na’ (24.3 miles)

173

(6.5) Mr. Sanford’s use of ’u- ‘DIST’ to describe comparatively long distance […] 175 nanaa ’uniit,

MED.across DIST.upriver.PUNC

(2.1) 176 fish camp.

(2.6) 177 Nanaa ’uniit,


(1.7) 178 Kedi eni Na’,

current.flows.in.REL.river

‘The fish camp across and [distantly] upriver at ‘creek that current flows in’ […]’


‘How We Went Hunting Out in the Country’, 00:06:25.950-00:06:35.400. Kari 2010:99))

6.5.1.2 Mrs. John s Use of the Distal Prefix u-

Mrs. John uses the distal prefix ’u- nineteen times in her narrative to refer to

eleven legs of the journey. Just as we saw with Mr. Sanford, the geographic

distances described with this prefix cover a large range, from less than one mile to

32 miles (mean 13.3 miles). Table 6.11 shows a summary of the legs of the journey

that Mrs. John describes with directionals prefixed with ’u-, followed again by

illustrative maps and texts of three legs.

174

Table 6.11: Journey Legs described by Mrs. John Using ’u- ‘DIST’


Leg Description

Length in miles


order

Tes T’aa Menn Tes (hill) to Taggos Menn’ (lake) 0.8

Ts’i ten’ Kats’etses Na’ (Pass Creek) to edidlende (Goat Creek)

4.4

Taggos Menn’ (lake) to Mendaesde (Mentasta village) 4.8

Sasluugu (Suslota) past Bes Ce’e 6.4

Natae de (Batzulnetas village) to Sasluugu (Suslota) along trails 7.3

Natae de (Batzulnetas village) to Stl’aa Caegge (mouth of Slana R.)

8.6

Natae de (Batzulnetas village) past Ts’abael K’edigha 9.3

Tsic’e ggodi Na’ (Jacksina R.) to Ts’i ten’ Kats’etses Na’ (Pass Creek)

17.3

Stl’aa Caegge (mouth of Slana R.), down Copper River to Tsiis Tl’edze Na’ (Chistochina R.)

23.8

Natae de (Batzulnetas village) to Copper River headwaters 32

Copper River headwaters to Natae de (Batzulnetas village) 32

The shortest leg in Mrs. John’s narrative described with a directional containing

’u- ‘DIST’ is shown in Map 6.9. In this leg, which is less than one mile in length,

Mrs. John describes traveling from Tes T’aa Menn’ Tes (hill near Fifteen Mile Lake)

to a lake known as Taggos Menn’ using ’unaa ‘DIST.across.PUNC’. This excerpt is

shown in (6.6).

175


Tes T’aa Menn’ Tes to Taggos Menn’ (<1 miles)

(6.6) Mrs. John’s use of ’u- ‘DIST’ to describe comparatively short distance 84 Tes T’aa Menn’ Tes,

hill.beneath.lake.hill

(1.2) 85 kaxatl-,

ABAN

86 xatl ’e kekude dze’ sled with, and 3PL.SUB. pl.go.IMPERF.against.a.place thus

’unaa, DIST.across.PUNC

87 Tes T’aa Menn’ Ts’edini eni yii, hill.beneath.lake.water.flow.from.REL 3S.NH

(1.2) 88 yii na’ ngge’ takake-,

3S.NH river upland ABAN

176

89 kakede . 3PL.SUB.pl.go.IMPERF.up

‘They would pass on sleds by ‘hill of lake beneath the hill’, [distantly] across where a stream ‘one that flows from ‘lake beneath the hill’ they would go upland there on that creek.’

(1.4) 90 Yihwts’en xunaann’ htede dze’,

there.AREA.from AREA.across 3PL.SUB.sp.go.IMPERF.INCEP thus

91 Taggos Menn’ ’ehdide dze’, swan.lake 3PL.SUB.pl.go.IMPERF thus

‘From there they would go across and they would come to ‘swan lake’.’

((Katie John, Natae Nenn’ ‘The Batzulnetas Country’, 00:03:11.280–00:03:26.790. Kari 2010:78))

Mrs. John uses the prefix ’u- to describe a journey leg of an intermediate length

in (6.7) below. Specifically, she uses ’utsiidze ‘DIST.downland.ABL’ to describe the

route from Tsic’e ggodi Na’ (the Jacksina River) to Ts’i ten’ Kats’etses Na’ (Pass

Creek). The route, which measures approximately 17.3 miles, is depicted in Map

6.10.

(6.7) Mrs. John’s use of ’u- ‘DIST’ to describe a comparatively intermediate distance 212 Yii Jack-,

3S.NH ABAN

(0.7) 213 tsighe-,

ABAN

214 Tsic’e ggodi Na’ tsen INDEF.SUB.rock.chip.CAUS.REL.river downland

ts’ets’ede . 1PL.SUB.pl.go.IMPERF.out

‘Jack...there at ‘rock is chipped’ creek’ we come out to the lowlands.’

(1.3) 215 Yets’e ’utsiidze yet,

there.from DIST.downland.ABL there

(2.4) 216 Ts’i ten’ Kats’etses Na’ keniide

bow.1PL.SUB.move.flexible.O.IMPERF.up.river 3PL.SUB.say.IMPERF.where

yet xu’aadze ta kanats’ede there area.other.side.from among 1PL.SUB.pl.go.IMPERF.up.ITER

177

dze’. thus

‘From there from [distantly] lowland (the Nabesna R.) to where they call ‘we lift up a bow creek’ (Pass Creek), we come back up from the other side.’

((Katie John, Natae Nenn’ ‘The Batzulnetas Country’, 00:07:57.800–00:08:10.620. Kari 2010:84-85))


Tsic’e ggodi Na’ to Ts’i ten Kats’etses Na’ (17.3 miles)

Finally, the two longest journey legs described by Mrs. John using ’u- are the

routes from Natae de (Batzulnetas village) to the Copper Glacier at the headwaters

of the Copper River and back again. This route is shown in Map 6.11, and the

178

corresponding text for the return trip is given in (6.8). Here she uses ’utsiit

‘DIST.downland.PUNC’ to describe the 32 mile journey.


Natae de to Copper Glacier and Back (32 miles)

(6.8) Mrs. John’s use of ’u- ‘DIST’ to describe a comparatively long distance 165 De xona,

ADVZ.TEMP.SPATIAL then

(1.9) 166 hwti -,

ABAN

167 hwte gguusi ts’en, weather.is.lukewarm.IMPERF.REL from

168 ta xona ts’enakede . among then 3PL.SUB.pl.go.IMPERF.out.ITER

(0.3) 169 Dze’ ’utsiit yet,

thus DIST.downland.PUNC there

179

(2.0) 170 deghak’ae ninakede .

3PL.REFL.POSS.home 3PL.SUB.pl.go.IMPERF.ITER.TERMIN

‘As it became spring there they would come back out [from the uplands around Copper Glacier] and they would stop again in the [distant] lowlands at their home [at Natae de].’


6.5.1.3 Mr. Tansy s Use of the Distal Prefix u-

We see the same pattern inn Mr. Tansy’s narrative, where he uses the distal

prefix ’u- nine times (note that his narrative is much shorter than that of Mr. Sanford

or Mrs. John), referring to five separate legs of the journey. The distances he refers

to with ’u- range from approximately 1.72 miles to approximately 7.2 miles (mean

4.4 miles), a smaller range than we observed for Mr. Sanford and Mrs. John, but still

relatively variable. Table 6.12 shows a summary of the legs of Mr. Tansy’s journey

that he describes with directionals prefixed with ’u-.

Table 6.12: Journey Legs described by Mr. Tansy Using ’u- ‘DIST’


Leg Description

Length in miles


order

Tl’ahwdicaax Na’ (Yanert Fork) to Nts’ezi Na’ (river) 1.7

Saas Nelbaay upstream to Saas Nelbaay Na’ Ngge’ (river uplands)

3

Taben ’aa Tayene’ (West Fork R. plain) to Taben ’aa Bene’ (lake)

4.7

Tanidzehi Deyii Na’ (river) through pass to Dghateni (trail) 5.2

Taben ’aa Bene’ (lake) across plain to Ben Datgge Na’ (river) 7.2

180

The shortest journey leg Mr. Tansy describes using a directional containing the

distal prefix ’u- is from Tl’ahwdicaax Na’ (Yanert Fork) to Nts’ezi Na’ (Nts’ezi

stream). This leg, which is approximately 1.7 miles long, is shown in Map 6.12, and

the corresponding text is shown in (6.9). He uses two directionals here containing ’u-

: ’udaa’a ‘DIST.downriver.ALL’ in IU 27, and ’unaa ‘DIST.across.PUNC’ in IU 28.


Tl’ahwdicaax Na’ to Nts’ezi Na’ (1.7 miles)

(6.9) Mr. Tansy’s use of ’u- ‘DIST’ to describe comparatively short distance 24 Ni denta u’,

sometimes, occasionally EVID

(0.7) 25 yet,

there

181

(0.8) 26 Tl’ahwdicaax Na’,

headwaters.be.valuable.river

(0.4) 27 ’udaa’a .

DIST.downriver.ALL

(0.4) 28 ’Unaa daa’a ts’its’ede dze’

DIST.across.PUNC downriver.ALL 1PL.SUB.pl.go.IMPERF.out thus

dae’ Nts’ezi Na’ hwts’e’, thus N.river AREA.to

(0.5) 29 tes ninats’ede .

hill 1PL.SUB.pl.go.IMPERF.TERMIN.ITER

‘Sometimes then there we come out [distantly] downstream and [distantly] across and downstream of ‘valuable headwaters stream’ (Yanert Fork), and that way we come back through a pass to Nts’ezi Stream.’

((Jake Tansy, Saen Tah Xay Tah C’a u’sghide

‘We Used to Travel Around in Summer and Winter’, 00:00:46.710-00:00:57.290. Kari 2010:61))

An example of an intermediate-length journey leg that Mr. Tansy describes using

a directional prefixed with the distal morpheme ’u- is the route from the wide, flat

river plain of Taben ’aa Tayene’ to the lake called Taben ’aa Bene’. This route is

about 4.7 miles long and is shown in Map 6.13. The corresponding text is shown in

(6.10). Mr. Tansy uses ’utsii ‘DIST.downland’ three times in this excerpt, in IUs 76,

77, and 78.

182


Taben ’aa Tayene’ to Taben ’aa Bene’ (4.7 miles)

(6.10) Mr. Tansy’s use of ’u- ‘DIST’ to describe comparatively intermediate distance 76 Ye u’ ’utsii ye c’a,

then DIST.downland there FOC

(0.3) 77 ’utsii ye Taben ’aa Tayene’,

DIST.downland there water.lake.current.flows.river.plain

78 ’utsii taz’aa de, DIST.downland ADVZ.TEMP.SPATIAL

(0.3) 79 Taben ’aa Bene’ hwdi’aan.

water.lake.current.flows.lake P.has.name.IMPERF.AREA

‘And [distantly] downland from there [distantly] downland of ‘lake current flows-river plain’ in the [distant] downland is situated the water called ‘lake current flows-lake’.’


00:02:31.900-00:02:38.230. Kari 2010:63))

183

The longest leg Mr. Tansy describes using a directional prefixed with ’u- is the

final leg of his narrated journey, from the lake called Taben ’aa Bene’ overland to

the stream called Ben Datgge Na’, near the village of C’ilaan Na’ (Valdez Creek).

This leg is shown in Map 6.14 and measures approximately 7.2 miles long. The

corresponding text is given in (6.11), in which Mr. Tansy uses ’utsiit

‘DIST.downland.PUNC’ to describe the route.


Taben ’aa Bene’ to Ben Datgge Na’ (7.2 miles)

(6.11) Mr. Tansy’s use of ’u- ‘DIST’ to describe comparatively long distance 82 Xona yet xu,

then there AREA

83 tsene yes kanats’ede dze’ downland.ALL dwarf.birch 1PL.SUB.pl.go.IMPERF.up.ITER thus

184

ye Ben Datgge, there lake.PROX.up.ALL

‘Then in the lowlands we ascend again in the dwarf birch (buckbrush), and’

(1.3) 84 cu yae’ ts’inats’ede dze’

also, else 1PL.SUB.pl.go.IMPERF.from.ITER thus

Ben Datgge Na’ hwtsene xona ye u’ lake.PROX.up.ALL.river AREA.downland.ALL then then

’utsiit Ts’itu’ gha. DIST.downland.PUNC major.river in.relation.to

‘there at ‘lake up above’ we come back out that way to ‘stream of lake up above’. In the [distant] lowlands then there by ‘major river’ (Susitna River).’


00:02:48.580-00:02:52.230. Kari 2010:63))

We have seen that for each of the three speaker’s, the range of distances covered

by their use of the distal deictic directional prefix ’u- is quite wide: between 1.5 and

24.3 miles for Mr. Sanford, between 0.8 and 32 miles for Mrs. John, and between 1.7

and 7.2 miles for Mr. Tansy. The next section discusses the way speakers use the

medial directional prefix na- to describe relative distances.

6.5.2 The Medial Prefix na-

The medial directional prefix na- occurs only five times in the entire corpus, and

was used only by Mr. Sanford.5 His uses of this prefix are discussed below; the

5 Mrs. John in fact does use the directional prefix na- ‘MED’ once, but her single

use of na- is not possible to analyze for distance. It occurs at the very end of her

narrative and is used to reference the direction from her actual location while she

was telling the travel narrative to a location called Nexk’aedi ‘lookout’. The excerpt

185

prediction based on the historical source of this prefix is that it will encode distances

that are comparatively shorter than those encoded by ’u- ‘DIST’.

in (i) is not part of her description of a migratory route, but rather part of a brief

explanation of the source of the toponym Nexk’aedi.

(i) Mrs. John’s use of na- ‘MED’ 337 Yii gha’ c’a Nexk’aedi

3S.NH because.of FOC lookout

xii’edi ’aan. 3PL.SUB.AREA.OBJ.give.P.name.PERF

‘That is why they call it ‘Lookout’.’

(1.1) 338 Nanuuxu yae’,

MED.upriver.AREA

(0.3) 339 K’e t’aenn t’aax yae’.

Mt. Sanford beneath

‘[Intermediately] upriver that way, it is beneath ‘K’e t’aenn’ (Mount Sanford).’


This is one of the few instances in the narratives when the speaker does not use

directional in a purely discourse-internal manner, but rather to orient the listener (in

this case, Jim Kari, who was recording the narrative) to the location under discussion

relative to the interlocutors’ location. Because Mrs. John narrative was recorded in a

moving car traveling from Mentasta to Slana (Kari 2010: 74), it is not possible to

map her use of na- ‘MED’ accurately here. Mr. Tansy does not use this directional

prefix at all.

186

6.5.2.1 Mr. Sanford s Use of the Medial Prefix na-

Unlike the distal prefix ’u-, which is used by speakers to describe a wide range of

distances, Mr. Sanford’s use of the medial prefix na- is restricted to directions and

locations that are only a short distance, usually across a river or stream. He uses this

morpheme in three episodes, summarized in Table 6.13. In each case, his use of na-

refers to distances that are less than one mile. Note that these uses do not describe

paths of travel but are instead descriptors of static locations.

Table 6.13: Distances Described by Mr. Sanford Using na- ‘MED’


Leg Description Length in miles (approximate)

Across Ts’itae Caegge (mouth of Sanford R.) to his childhood home

<1

Fish camp at Ts’itu’ K’et, across from Kedi eni Na’ <1

Snuu Caegge (Sinona R. mouth), across from mouth of Tsidghaazi Na’ (river)

<1

The first espisode is a description of his childhood home as being ‘intermediately

across’ from Ts’itae Caegge (mouth of Sanford River). This is shown in Map 6.15,

with the corresponding text given in (6.12).

187

Map 6.15: Example of na- ‘MED’ for the Area Across

from Ts’itae Caegge (<1 mile)

(6.12) Mr. Sanford’s use of na- ‘MED’ 07 Ts’itae Caegge,

river.flows.straight.rivermouth

(1.0) 08 Ts’itae Caegge nanaa neghak’ae,

river.flows.straight.rivermouth MED.across 1PL.POSS.home

‘[Intermediately] across from ‘mouth of flows straight’, ‘mouth of flows straight’ (was) our home.’

09 Stsucde ’iin ye hdaghalts’e’, 1S.POSS.grandmother PL.HUM there 3PL.SUB.pl.sit/stay.PERF

‘My grandmothers lived there.’

188

(1.8) 10 stsucde ’iin ye hdaghalts’e’.

1S.POSS.grandmother PL.HUM there 3PL.SUB.pl.sit/stay.PERF

‘My grandmothers lived there.’


00:00:27.400-00:00:35.420. Kari 2010:92))

The next two episodes in which Mr. Sanford uses directionals prefixed with na-

‘MED’ are two separate discussions about the same location, the region where the

Chistochina River meets the Copper River. Two smaller creeks, Tsidghaazi Na’ and

Kedi eni Na’ also meet the Copper River here, and Mr. Sanford’s fish camp is

located at Ts’itu’ K’et in this area as well. Mr. Sanford discusses this area twice. The

first is shown in Map 6.16 and (6.13), in which Mr. Sanford uses nanaa

‘MED.across’ twice to describe the spatial relationship between his fish camp at

Ts’itu’ K’et and Kedi eni Na’.

189

Map 6.16: Example of na- ‘MED’ for the Relationship Between

Ts’itu’ K’et and Kedi eni Na’ (<1 mile)

(6.13) Mr. Sanford’s use of na- ‘MED’ 175 nanaa ’uniit,


(2.1) 176 fish camp.

(2.6) 177 Nanaa ’uniit,


190

(1.7) 178 Kedi eni Na’,

current.flows.in.REL.river

‘The fish camp [intermediately] across and upriver at ‘creek that current flows in’[…]’


00:06:25.950-00:06:35.400. Kari 2010:99))

The second espisode in which Mr. Sanford uses the medial prefix na- to describe

locations in this region is shown in Map 6.17. In the corresponding text in (6.14),

uses nanaat ‘MED.across.PUNC’ and nanuughe ‘MED.upriver.AREA’ to describe the

spatial relationship between the mouth of Tsidghaazi Na’ and Snuu Caegge (Sinona

Creek mouth).

191

Map 6.17: Example of na- ‘MED’ for the Relationship Between

Tsidghaazi Na’ and Snuu Caegge (<1 mile)

(6.14) Mr. Sanford’s use of na- ‘MED’ 205 Yihwts’en xona,

there.AREA.from then

(0.4) 206 Tsidghaazi Na’ tggase’.

rock.in.rough.REL.river up.ABL

(1.8) 207 Na’stetnaes.

1PL.SUB.travel.nomadically.IMPERF.ITER

‘From there we would move back from above at ‘rough rock creek’ (Boulder Creek).’

(2.3) 208 Nanaat,

MED.across.PUNC

192

209 nanuughe Snuu Caek’e MED.upriver.AREA brush.rivermouth

ts’inats’etnaes. 1PL.SUB.move.nomadically.IMPERF.from.ITER

‘We came out [intermediately] across in upriver area of ‘brushy mouth’ (Sinona Creek mouth).’

((Adam Sanford, C’uka Ts’ul’aen’i gha Nen’ Ta’stede dze’ ‘How We Went Hunting Out in the Country’, 00:07:42.180-00:07:51.940. Kari 2010:102))

In summary, Mr. Sanford’s use of the medial directional prefix na- is restricted

to descriptions of very short distances, and furthermore these do not describe paths

of travel, but rather static locations of referents in the landscape. The next section

examines the use of the proximal directional prefix da-.

6.5.3 The Proximal Prefix da-

Examples of Mr. Sanford’s and Mrs. John’s usages of the proximal prefix da- are

discussed in turn below. Again, the prediction for this morpheme is that it will

encode distances that are comparatively shorter than both the distal prefix and the

medial prefix. Mr. Tansy’s single use of this morpheme is not possible to map and is

not discussed.

6.5.3.1 Mr. Sanford s Use of the Proximal Prefix da-

Mr. Sanford uses a directional prefixed with da- ‘PROX’ only once in his

narrative, to describe a leg of his journey that begins in the sheep hunting grounds

across from the mouth of the river called Una’ Hwnelk’ezi Na’ back down to Ts’itae

Na’ (Sanford River). This leg measures approximately 3.6 miles and is shown in

Map 6.18, with the corresponding text given in (6.15). Notice that this distance is

193

longer than some of those described with ’u- ‘DIST’. In fact, interestingly, this is the

return along a path that Mr. Sanford described in the previous discourse with ’unaa

‘DIST.across.PUNC’.

Map 6.18: Example of da- ‘PROX’ for the Route from

Una’ Hwnelk’ezi Na’ to Ts’itae Na’ (<1 mile)

(6.15) Mr. Sanford’s use of da- ‘PROX’ 85 Da- danii ts’en nggase’,

ABAN MED.upriver from upland.ABL

(1.0) 86 tat’aa nggase’,

valley upland.ABL

87 ’adii na’stetnaes. now 1PL.SUB.travel.nomadically.IMPERF.ITER

‘From [nearby] upriver, from the uplands to the valley from the uplands we would return.’


00:03:12.790-00:03:17.100. Kari 2010:95))

194

6.5.3.2 Mrs. John s Use of the Proximal Prefix da-

Mrs. John uses the proximal prefix da- eleven times during her narrative,

referring to seven individual legs of her described journey. The geographic distances

she refers to with this prefix range from approximately 0.9 miles to 17.3 miles (mean

6.6 miles). Table 6.14 summarizes the legs of Mrs. John’s journey that she describes

using directionals prefixed with da- ‘PROX’. Again, we see that some of these

distances are in fact longer than some of the ones she describes with ’u- ‘DIST’.

Table 6.14: Journey Legs Described by Mrs. John Using da- ‘PROX’

and their Length in Miles

Leg Description

Length in miles


order

Downland from Kolgiisde into hunting grounds 0.9

Upland from Kolgiisde into hunting grounds 1.4

Mendaes Caegge (river mouth) to Kolgiisde 3.5

From Tl’ogh Tnge naay to Bes Ce’e 5.2

From Bes Ce’e to Stl’aa Caegge (Slana) 8

From edidlende into the uplands, to approximately Men Ni ga’aa Delyaade (lakes)

10

From Natae de (Batzulnetas village) to Tanaade Menn’ (Tanada lake)

17.3

Among the shorter distances described by Mrs. John using da- is shown in Map

6.19, with the corresponding text in (6.16). Mrs. John uses danggex

‘PROX.upland.AREA’ to describe the spatial relationship between the place known as

Kolgiisde and the nearby upland muskrat hunting grounds.

195


Kolgiisde to Nearby Uplands (<1 mile)

(6.16) Mrs. John’s use of da- ‘PROX’ to describe comparatively short distance 12 Yehwts’en danggex dze’ Mendaes hwt’aenn

there.area.from PROX.upland.AREA thus shallows.lake everyone

’iinn (Hx), PL.HUM

(0.8) 13 uxac’ehdelyiis dzen ’uka.

3PL.SUB.3INDEF.OBJ.weapon.go.PERF.PERAMB muskrat 3S.OBJ.in.quest.of

‘From there from the [nearby] uplands the ‘shallows lake’ people used to hunt for muskrat’.


Mrs. John also uses the proximal morpheme da- to describe longer distances. An

example of this is shown in Map 6.20, with the corresponding text found in (6.17).

Mrs. John uses datsen ‘PROX.downland’ to describe the 8 mile overland journey

between Bes Ce’e and Stl’aa Caegge (the mouth of the Slana River).

196


Bes Ce’e to Stl’aa Caegge (8 miles)

(6.17) Mrs. John’s use of da- ‘PROX’ to describe comparatively intermediate distance 260 Xona,

then

(1.0) 261 yi ’e ta xona da-,

DEM with, and among then ABAN

262 datsen, PROX.downland

(1.0) 263 Stl’aa Caegge dze’,

rear.rivermouth thus

264 tsene ta, downland.ALL among

(0.6) 265 xii’e tade dze’ ’utsii

that.with 3PL.SUB.pl.go.IMPERF.INCEP thus DIST.downland

Stl’aa Caegge hneye dze’. rear.rivermouth 3PL.SUB.camp.IMPERF thus

‘Then there [e.g., Bes Ce’e] to the [nearby] lowlands at ‘mouth of rear river’ (mouth of Slana River) they went with that to the

197

lowlands. They would camp downland by ‘mouth of rear river’ […]’


Mrs. John also uses directionals prefixed with da- ‘PROX’ to describe even longer

legs of her journey. The longest distance she describes with this morpheme measures

approximately 17.3 miles in length, from Natae de (Batzulnetas village) to Tanaade

Menn’ (Tanada Lake). As we have seen from (6.6)-(6.8) this distance is strikingly

longer than many of distances she describes with the distal prefix. This route is

shown in Map 6.21. In the corresponding text in (6.18), Mrs. John uses the

directional danggeh ‘PROX.upland.AREA’.


Natae de to Tanaade Menn’ (17.3 miles)

198

(6.18) Mrs. John’s use of da- ‘PROX’ to describe comparatively long distance 171 Naenn du’,

1PL

172 kanii hwts’en danggeh, ADJACENT.upriver from PROX.upland.AREA

173 Tanaade Menn’ xu kenii xuh tah moving.water.where.lake AREA 3PL.SUB.say.IMPERF AREA among

nenn’. earth

174 Nenenn’ ta kughile’ dze’, 1PL.POSS.earth among clf.pl.obj.IMPERF thus

175 kuh tah. area among

(0.7) 176 Nin’ta u’stedii .

earth.among 1PL.SUB.pl.go.IMPERF.PERAMB

‘As for us, the next place upriver from there, [nearby] upland they call ‘moving water lake’ (Tanada Lake). Within that area was our country and within there we would go around the country.’


6.6 Discussion: Lexicalization in the Prefixes

Table 6.15 provides a summary of the ranges of distances that each speaker

describes with each of the three deictic prefixes, as discussed in Section 6.5 above.

Table 6.15: Summary of Ranges in Miles of Distances Described

by Speaker and Directional Prefix

Mr. Sanford (narrative route

length 152 miles)

Mrs. John (narrative route

length 254 miles)

Mr. Tansy (narrative route

length 107 miles)

Total range covered by

prefix

’u-

DIST 1.5-24.3 (14 legs)

0.8-32 (11 legs)

1.7-7.2 (5 legs)

0.8-24.3

na- MED

<1 (3 regions)

- - -

da-

PROX <1

(1 leg) 0.9-17.3 (7 legs)

- <1-17.3

199

We can see that there is no clear distal vs. medial vs. proximal distinction among

the use of the prefixes, either in terms of individual speakers or in terms of

individual prefixes. As to the first, Mr. Sanford’s use of ’u- ‘DIST’ covers a wide

spatial range, and his uses of na- ‘MED’ and da- ‘PROX’ do not show any difference,

as these are both used to describe small distances of less than one mile only. Mrs.

John also does not show a clear three-way distinction in her uses of the prefixes. Her

uses of ’u- ‘DIST’ cover a wide range of distances, from less than one mile to

approximately 32 miles. Her use of the proximal morpheme da- covers a range that

at the low end is similar to the low end of her uses of the distal morpheme, and at the

high end is more than half of the total range covered by the distal morpheme. Mr.

Tansy restricts himself to the distal morpheme only. Thus none of the three speakers

exhibit a clear three-way semantic distinction between the deictic directional

prefixes.

In terms of the use of the prefixes across all three speakers, again a three-way

distinction fails. The distal and proximal prefix can both be used to describe

relatively short distances, and while the upper end of the range covered by the distal

morpheme is higher than that covered by the proximal morpheme, it is only larger by

approximately one-quarter. Note also that the medial prefix is hardly selected at all,

only by Mr. Sanford and only to describe very short distances of considerably less

than one mile, and not for describing a path of travel.

These findings support the idea that the historical semantic strength of the Proto-

Athabascan deictic demonstratives has weakened considerably in actual usage, and

speakers are not relying on them heavily to distinguish between locations found at

200

nearby locations versus those found far away versus those found somewhere in

between.6 A speaker’s choice to make such a distinction is of course relative, and the

distance from point A to point B may be best described as ‘near’ under one set of

circumstance and as ‘far’ under another. However, even within individual travel

narratives, in which we can presume the speaker has a mental map of the entire route

he or she is describing, the deictic prefixes are not doing as much of the work of

distinguishing distances as we might assume. The prefixes have undergone both

phonological fusion and semantic bleaching, and thus we can say that they are

lexicalizing, on their way to forming a set of single semantic units with the

directional stem.

This does not, however, answer the question presented in Section 6.1, that of

how speakers use directional morphology to distinguish between relative distances in

their travel narratives. The next chapter shows that speakers are using the directional

suffixes (i.e., the allative, ablative, punctual, and area suffixes), rather than the

6 As with my discussion of Mr. Pete’s gradual semantic shift in previous

chapters, this is not to imply that Mr. Sanford, Mrs. John, or Mr. Tansy are in any way deficient speakers of Ahtna. To the contrary, they are all fully fluent in the language. Furthermore, these travel narratives showcase their expertise on land use and the geography of the Ahtna region. As Kari rightly points out, these stories are remarkable for being “very rare, spontaneously offered texts about incredibly hazardous travel to high mountains” and display an “array of tactical and linguistic feats being presented by [the speakers]. If we want to emphasize these prefixes are not so precise as [previous analyses] claim they are, that is really a tiny piece of the powerful message of the rare text” (Kari, p.c.). I concur wholeheartedly, and wish to restate here that the present discussion is an analysis of the morphology only, and is not meant to be a commentary on the extraordinary content of the narratives.

201

deictic prefixes, to structure their discourse in ways that indicate differences in

distance.

202

Chapter 7

Directional Suffixes and Discourse Structure

7.1 Introduction

The previous chapter looked at the way directional prefixes pattern in the

spontaneous discourse of the Ahtna travel narratives. It was found that speakers are

not fully exploiting the presumed deictic semantics of the prefixes to make clear

distinctions between near vs. intermediate vs. far distances. The present chapter turns

to how speakers employ the directional suffixes in spontaneous speech. The data for

this chapter are the same three narratives and Kari’s (2008) database of Ahtna

toponyms described in the previous chapter.

Section 7.2 below examines the question of whether the suffixes, like the

directional prefixes, may be undergoing the semantic bleaching that is part of

lexicalization, and finds that there is no evidence of a loss of semantic clarity.

Section 7.3 then discusses the role the suffixes – in particular the punctual and area

suffixes – play in structuring the larger discourse of the travel narrative into (a) legs

of the physical journey and (b) episodes within the narrative. This discourse use of

the suffixes could not have been predicted from the literal interpretation of the

suffixes alone. Section 7.4 then argues for the hypothesis that speakers are then

repurposing the discourse function of the suffixes to make up for the deictic

203

distinctions that are bleaching from the prefixes. Section 7.5 contains concluding

remarks.

7.2 Literal Use of the Directional Suffixes

This chapter discusses the directional suffixes; for convenience, the tripartite

morphemic structure of the directions is reproduced in Table 7.1 below.


(from Kari 1990)










nse’ ‘ahead’

The travel narrative data do not reveal any evidence that the directional suffixes

are undergoing lexicalization like the directional prefixes. While they have fused

phonologically to the lexical stem, they do not display any of the semantic bleaching

that is found in the prefixes. The suffixes in (7.1)-(7.4) below show no difference

between their lexical meaning and their interpretation in discourse.

7.2.1 The Punctual Suffix -t

An example of the use of the punctual suffix -t is in (7.1) below.

204

(7.1) Literal use of punctual directional suffix -t 180 Gha yet ta,

in.relation.to there among

(1.6) 181 ts’eneye ts’en ye xanggat,

1PL.SUB.camp.IMPERF from there ADJACENT.upland.PUNC

‘[by] there, we would camp at the next place upland.’


In this excerpt Mrs. John is describing travel in the Tanada Lake area. She uses

xanggat ‘ADJACENT.upland.PUNC’ to pinpoint the travelers’ campsite as ‘the next

place upland’ (i.e., the next campsite upland from Tanada Lake).

7.2.2 The Area Auffix -xu

An example of the use of the area suffix -xu is in (7.2) below.

(7.2) Literal use of area directional suffix -xu 193 Ye naak’e u’stedii xu debae

there mineral.lick 1PL.SUB.pl.go.IMPERF.CUST.PERAMB AREA sheep

gha. in.relation.to

‘We would go around on mineral licks for sheep.’

(1.8) 194 Kude diye dze’,

(distance.)be.short thus

(0.6) 195 debae c’ilaen you know.

sheep be, be born

‘Sheep are nearby, you know.’

(0.8) 196 Kude diyede k’a ts’eghaax.

(distance.)be.short.where EMPH 1PL.SUB.kill.pl.O.IMPERF

‘We would kill them right nearby.’

(0.9) 197 Naxae ts’eldii i dze’ ukae k’a

1PL.SUB.pack.pl.go.IMPERF.ITER.REL thus 3S.OBJ.by.means.of EMPH

ts’edelts’iix. 1PL.SUB.sit/stay.IMPERF

‘We packed them back. We just lived on that.’

205

(1.1) 198 Udzih cu ’utggu utak’a.

caribou also, else DIST.up.AREA interspersed

‘Caribou also [distantly up] above [in an area] there were interspersed.’

((Adam Sanford, C’uka Ts’ul’aen’i gha Nen’ Ta’stede dze’ ‘How We Went Hunting Out in the Country’, 00:07:06.700-00:07:21.070. Kari 2010:100))

Here Mr. Sanford is describing land use in the area around Kateni Na’ Ngge’.

This portion of the story takes place outside of the trajectory of the narrative; that is,

he is providing background information on this particular stopping point in the

journey before resuming travel. In these lines he is describing the wanderings of the

hunters around the area, and in IU 198 he uses ’uttgu ‘DIST.up.AREA’ to describe the

distribution of caribou through the general area.

7.2.3 The Allative Suffix -e

An example of the use of the allative suffix -e is in (7.3).

(7.3) Literal use of allative directional suffix -e 32 Yihwts’en xona Natii Na’ Ngge’,

there.area.from then N.river.upland

(2.2) 33 xona ’utgge yii,

then DIST.up.ALL 3S.NH

34 xungge’ tah kude diye, area.upland among (distance.)be.short

‘From there, then in ‘natii river uplands’, then [to distantly] upward, upland there is a short distance.’


00:01:18.860-00:01:24.970. Kari 2010:93))

206

Here Mr. Sanford is discussing travel to the uplands of the river known as Natii

Na’. In IU 33 he uses the allative suffix in ’utgge ‘DIST.up.ALL’ to indicate the

direction ‘upward’.

7.2.4 The Ablative Suffix -dze

An example of the usage of the ablative suffix -dze is in (7.4).

(7.4) Literal use of ablative directional suffix -dze 16 Ba’aadze den u’,

outside.ABL ADVZ.TEMP.SPATIAL EVID

17 Tl’ahwdicaax Na’, headwaters.be.valuable.river

18 hwts’e’ hwcets’ede de ena’udghidlen, AREA.to 1PL.SUB.base.pl.go.IMPERF.AREA.where streams.join.together

19 xunt’ae. AREA.SUB.be.thus.IMPERF

‘[from] over there to ‘valuable headwaters stream’ (Yanert Fork) we ascend what is ‘streams join together’ (Louise Creek).’


00:00:34.330-00:00:38.950. Kari 2010:60))

In this excerpt Mr. Tansy had been discussing the various routes one might take

to reach the uplands of the river known as Tl’ahwdicaax Na’, including one route via

a mountain pass at a place known as Dghateni. In IU 16 he uses ba’aadze

‘outside.ABL’ to indicate the direction from Dghateni to Tl’ahwdicaax Na’.

It must be noted that while the examples above show that the meaning of the

suffixes is not bleaching in the same way as that of the prefixes, this is due in part to

the non-paradigmatic nature of the suffixes. The prefixes are scalar and are thus

susceptible to a leveling of the cline between them, but the suffixes are composed of

two pairs of semantic opposites (areal vs. punctual; allative vs. ablative) that occupy

207

the same suffix position following the directional stem. However, this does not mean

that their function in discourse can be fully predicted based on their lexical meaning

alone, as is shown in the following sections.

7.3 Discourse Use of the Directional Suffixes

We now to turn to two of the four suffixes discussed above, the punctual suffix

and the area suffix In addition to employing the directional suffixes for their literal

meanings in the spontaneous discourse of the travel narratives, speakers are also

recruiting these suffixes to structure the larger discourse.

7.3.1 Discourse Use of the Punctual Suffix -t

First let us examine the discourse use of the punctual suffix -t. The travel

narratives display an overwhelming co-occurrence of directionals containing the

punctual suffix on the one hand, and toponyms or definite locational nouns or

pronouns on the other. While not every toponym is preceded by a directional

containing the punctual suffix, nearly every instance of a directional containing the

punctual suffix is followed by a toponym. Of the 34 such directionals in the data, 28

(82.3%) of them occur within one IU of a toponym or locational noun, with most of

them immediately preceding the toponym or directional noun. Some examples are in

(7.5).

208

(7.5) Directionals with punctual suffix -t co-occurring with toponyms and locational nouns

1

(a) 68 ’Unggat uu gha,

DIST.upland.PUNC glacier in.relation.to

(0.2) 69 na-,

ABAN

(4.0) 70 yet k’a ’sdelts’iix,

there EMPH 1PL.SUB.pl.sit/stay.IMPERF

‘At an upland place there we stayed right by the glacier.’


00:02:34.880-00:02:41.490. Kari 2010:95))

(b) 133 Kaniit,

ADJACENT.upriver.PUNC

(1.0) 134 Una’ Tuu Koley Na’,

3S.POSS.river.water.EXIST.NEG.river (tpnm)

(1.3) 135 yehwk’e nats’etnaes.

like.that 1PL.SUB.travel.nomadically.IMPERF.ITER

‘[At] the next place upriver we went back to ‘its creek has no water creek’.’


00:05:01.300-00:05:06.240. Kari 2010:97))

(c) 37 ’Utsii Natae de hwts’en

DIST.downland.PUNC roasted.salmon.place (tpnm ) AREA.from

k’a, EMPH

1 Note that in rapid speech, the -t alternates with zero; the directional stem vowel

change remains.

209

38 ’utsiidze tene kulaen dze’. DIST.downland.ABL trail, road EXIST thus

‘From downland at ‘roasted salmon place’ (Batzulnetas Village) there are trails coming up from the lowlands.’


(d) 73 ’Unggat,

DIST.upland.PUNC

(0.2) 74 Xoos Ghadl Zdlaa keniide ye

horse.wagon.exist (tpnm ) 3PL.SUB.say.IMPERF.WHERE there

danggasts’en ta, PROX.upland.SIDE among

(0.6) 75 xona danggehdze dae’

then PROX.upland.FROM thus

tic’akede . 3PL.SUB.pl.go.IMPERF.out.into.the.woods

‘Up at where they call it ‘horse wagons are there’ there then they would go out into the country to the upland side or from the uplands.’


(e) 76 Ye u’ ’utsii ye c’a,

then DIST.downland.PUNC there FOC

(0.3) 77 ’utsii ye Taben ’aa Tayene’,

dist.downland.PUNC there water.lake.current.flows.river.plain (tpmn)

‘And downland from there downland of ‘lake current flows-river plain’ […].’


00:02:31.990-00:02:34.550. Kari 2010:63))

Speakers are using the punctual suffix (-t or its rapid-speech zero allomorph) on

the directionals to mark definite, identifiable locations, areas in the geography that

are bounded spaces, defined by their toponym or the edges of the physical structure

(e.g., a glacier). While it is not surprising that speakers would refer to such places in

210

a punctual manner, the boundary-defining nature of the punctual suffix plays a large

role in the structuring of the overall discourse. Recall from Chapter 5 that the travel

narratives are structured into episodes corresponding to legs of the physical journey,

in which speakers generally name a location, provide background information or

commentary (e.g., land use), and then move to the next location. Here we see the

suffix being used to define legs of the journey, stopping points at which the speaker

has the opportunity to discuss background information about each site before moving

on. As was seen in Chapter 5, the speakers then use various discourse markers (e.g.,

xona ‘then’ or yihwts’en ‘there.AREA.from’) to move the discourse forward to the

next episode or journey leg.

7.3.2 Discourse Use of the Area Suffix -xu

The area suffix -xu (and its allomorph -eh) also has a discourse function that

parallels that of the punctual suffix. Where the punctual suffix -t is described above

as segmenting both the physical journey into legs and the discourse into brief

episodes, the area suffix -xu is used to demarcate larger legs of the physical journey

and major episodes of discourse that are composed of many smaller episodes.

Example (7.6) shows the area suffix being used in this way. The excerpt starts at

the end of one leg of Mrs. John’s journey, a location called edidlende where she

describes hunting for sheep in IUs 190-192. The next leg of the journey begins in IU

193 with yihwts’en xona ‘from there then’. Note the use of the area suffix in

danggeh ‘PROX.upland.AREA’ in IU 193.

211

(7.6) Major and minor journey legs and discourse episodes, Mrs. John 190 edidlende yet ta xona,

streams.join.where (tpnm) there among then

(1.8) 191 debae keghiix dze’.

sheep 3PL.SUB.kill.pl.O.IMPERF thus

192 Yii debae ’e yet ’sdelts’iix. 3S.NH sheep with, and there 1PL.SUB.pl.sit/stay.IMPERF

‘They killed sheep there then at ‘where streams join’. We stayed there with that sheep.’

(0.8) 193 Yihwts’en xona danggeh ta,

there.area.from then PROX.upland.AREA among

(1.6) 194 c’ena’,

INDEF.POSS.river

(1.0) 195 ngge’ ta kets’ede dze’ yi

upland among 1PL.SUB.pl.go.IMPERF.AGAINST.A.PLACE thus DEM

ya xungge’, AREA.upland

(1.6) 196 kecdilaa de.

INDEF.clf.pl.O.AREA ADVZ.TEMP.SPATIAL

‘From there then in the uplands, as we go on upland, and there are names in the uplands there.’

(1.9) 197 Ts’akae,

woman

(0.3) 198 Ts’akae Ggan Nats’i baa

woman.small.1PL.SUB.move.attached.to.rope.PERF.CAUS.down (tpnm)

keniide ye, 3PL.SUB.say.IMPERF.where there

199 xuh kets’ude dze’. AREA 1PL.SUB.pl.go.IMPERF.against.a.place thus

(0.7) 200 Ye xanggat da men,

there ADJACENT.upland.PUNC POL.Q lake

(1.8) 201 Men Ni gha’aa Delyaade,

lakes.connected.together (tpnm)

‘Where they call ‘the thin lady was lowered on a rope’, they pass by there and the lakes up there, ‘lakes connected together’.’

202 gha ye xu’enn’ cu in.relation.to there area.other.side also, else

212

kats’ede dze’. 1PL.SUB.pl.go.IMPERF.up thus

‘There we go up over to the other side also (to the Nabesna R. drainage).’

(1.3) 203 Ye xu’en xona,

there area.other.side then

204 Tsi-, ABAN

205 Tsic’e ggodi Tl’aa INDEF.SUB.chip.CAUS.rock.REL.headwater (tpnm)

keniide, 3PL.SUB.say.IMPERF.where

206 ye kets’ede . there 1PL.SUB.pl.go.IMPERF.against.a.place

‘There over to ‘rock is chipped headwaters’ (Jacksina Creek headwaters) we came to there.’

207 Tsic’e ggodi keniide INDEF.SUB.rock.chip.CAUS.REL (tpnm) 3PL.SUB.say.IMPERF.where

de, ADVZ.TEMP.SPATIAL

208 ’adii du’, now

209 Jacksina, J.

(0.3) 210 Nondlae ’iinn,

white.person PL.HUM

(0.6) 211 Jacksina kehdi ’aan.

J. 3PL.SUB.give.P.name.PERF.AREA

‘Where they now call ‘rock is chipped’ the white people have named the place Jacksina.’

(1.1) 212 Yii Jack-,

3S.NH ABAN

(0.7) 213 tsighe-,

ABAN

214 Tsic’e ggodi Na’ tsen INDEF.SUB.rock.chip.CAUS.REL.river (tpnm) downland

ts’ets’ede . 1PL.SUB.pl.go.IMPERF.out

‘Jack...there at ‘rock is chipped’ creek’ we come out to the lowlands.’

213

(1.3) 215 Yets’e ’utsiidze yet,

there.from DIST.downland.ABL there

‘From there, from the lowlands […]’

((Katie John, Natae Nenn’ ‘The Batzulnetas Country’, 00:07:13.650–00:08:04.280. Kari 2010:84-85))

In IU 193, Mrs. John uses the directional danggeh ‘PROX.upland.AREA’ to move

the journey into the uplands above edidlende. However, she is not simply preparing

to describe the next upland location; rather, she is planning to discuss all the names

in the entire upland area, which she indicates in IUs 195-196 by saying yi ya xungge’

kecdilaa de ‘there are names in the upland there.’ Using danggeh

‘PROX.upland.AREA’ in IU 193 serves to let the listener know that the speaker has

the larger region in mind, not just the next single location.

The next 25 IUs consist of Mrs. John listing, in order, the named locations in the

uplands: Ts’akae Ggan Nats’i baa (IU 198), Men Ni gha’aa Delyade (IU 201),

Tsic’e ggodi Tl’aa (IU 205), and Ts’ice ggodi (IU 207). This leg of the journey ends

in IU 214 at Ts’ice ggodi Na’, leaving the uplands and heading into the lowlands

(tsen ‘downland’). The next journey leg begins in IU 215 with yets’e… ‘there.from’

(a variant of the more common yihwts’en ‘there.AREA.from’).

This discourse use of the area suffix in IU 193 also has an allative connotation.

When combined with yihwts’en xona ‘from there then’ in IU 192, the area suffix on

danggeh ‘PROX.upland.AREA’ tells the listener that the eventual destination is the

upland region although no overt allative marker is used. This demarcates the entire

upland region as the next major leg of the journey, which contains a number of

smaller legs, from one individually named location to the next (shown in Map 7.1

214

below). At the same time, this use of the area suffix also demarcates the following

discourse as a major episode, uniting several small episodes into one larger discourse

structure.

Map 7.1: Major and Minor Journey Legs in (7.6)

7.3.3 Discourse Patterning of Both Suffixes

Speakers also have the option of combining the two discourse uses of the

punctual and area suffixes described above. In one striking example, Mr. Sanford

uses the area suffix to demarcate the major journey leg he is about to describe with a

major discourse episode, and then uses the punctual suffix to mark the smaller

journey legs and minor discourse episodes. This is shown in (7.7). In this excerpt,

Mr. Sanford is describing the journey from Ben Tah (mentioned 18 IUs earlier) all

215

the way to the region above the timberline near Tsidghaazi Tl’aa on the ridge of

Mount Sanford.

(7.7) Major and minor journey legs and discourse episodes, Mr. Sanford 130 Du’, 131 yehwts’en,

there.area.from

(0.8) 132 ye nae’ ts’enaes ’utggu

there upriver 1PL.SUB.travel.nomadically.IMPERF DIST.up.AREA

daaghe nae’ k’e. timberline upriver on

‘From there we started out upstream to above the timberline on the upstream.’

133 Kaniit, ADJACENT.upriver.PUNC

(1.0) 134 Una’ Tuu Koley Na’,

3S.POSS.river.water.EXIST.NEG.river (tpnm)

(1.3) 135 yehwk’e nats’etnaes.

like.that 1PL.SUB.travel.nomadically.IMPERF.ITER

‘The next place upriver we went back to ‘its creek has no water creek’.’

[11 IUs about Una’ Tuu Koley Na’ omitted] 148 Duu yehwts’en xona ’uniit,

there.AREA.from then DIST.upriver.PUNC

(0.7) 149 Saas Dze k’et,

sand.mountain (tpnm) on

(0.4) 150 ninats’etnaes.

1PL.SUB.move.nomadically.IMPERF.TERMIN.ITER

‘From there, then we would stop again upriver at ‘sand mountain’.’

151 ’Utggat tes ghak’aay. DIST.up.PUNC hill mountain.ridge

(1.5) 152 ((COUGH)) Yet ninats’etnaes.

there 1PL.SUB.move.nomadically.IMPERF.TERMIN.ITER

‘Up on the flank of a hill we would stop.’

153 Tseles. ground.squirrel

216

(0.8) 154 Tseles gha yet

ground.squirrel in.relation.to there

nits’enaes. 1PL.SUB.move.nomadically.IMPERF.TERMIN

‘Ground squirrels, we stopped there for ground squirrels.’

[19 IUs about hunting ground squirrels at the hill at Saas Dze omitted] 173 Du’, 174 xona Tsidghaazi Tl’aa,

then rock.that.is.rough.headwaters (tpnm)

‘Then to ‘rough rock headwaters’.’


00:04:56.410-00:06:24.340. Kari 2010:97-99))

In IU 132, Mr. Sanford uses ’utggu daaghe ‘DIST.up.AREA timberline’ to

indicate that the next portion of the journey – the major journey leg and the major

discourse episode – will eventually reach the higher elevations above the timberline.

Again, note the allative function here: combined with the yehwts’en

‘there.AREA.from’ in IU 131, the sense is that the travel is moving toward the

timberline, although no allative morpheme or postposition has been used. These first

IUs serve to demarcate the next major leg of the journey, as well as the next major

episode in the discourse (i.e., the trip to above the timberline at Tsidghaazi Tl’aa).

Mr. Sanford follows this demarcation of the major leg with the first minor leg,

from the region around Ben Tah to Una’ Tuu Koley Na’. Note that this place name is

introduced with a directional containing the punctual morpheme -t in kaniit

‘ADJACENT.upriver.PUNC’ in IU 133. He then talks about caribou hunting at Una’

Tuu Koley Na’ for eleven IUs before moving to the next minor leg, the journey to

Saas Dze . This location is also marked with a directional containing the punctual

suffix, ’uniit ‘DIST.area.PUNC’ in IU 148. He then pinpoints his location at Saas Dze

217

even more finely in IUs 151 and 152 as being ’utggat tes ghak’aay ‘DIST.up.PUNC’

hill mountain ridge’ or ‘up on the flank of a hill’. After 21 IUs about hunting ground

squirrels around Saas Dze , the next leg is to Tsidghaazi Tl’aa, the region above the

timberline on the ridge of Mount Sanford. The major leg and its subdivisions into

smaller legs are shown in Map 7.2.

Map 7.2: Major and Minor Journey Legs in (7.7)

Figure 7.1 is a diagram of the major and minor episodes in the discourse showing

the suffixes used to name each location.

218

Figure 7.1: Major and Minor Discourse Episodes in (7.6)

In this analysis, we can finally make sense of (6.1) from the previous chapter,

reprinted below as (7.8). The region is depicted below the example in Map 7.3.

(7.8) Use of directionals in discourse (repeated from (6.1) 46 Du’ yihwts’en,

there.area.from

(0.2) 47 ts’inats’ede dze’ ’unggeh,

1PL.SUB.pl.go.IMPERF.from.ITER thus DIST.upland.ALL

‘From there, then we would start out to uplands.’

(0.8) 48 ’utggu daagha ngge’,

DIST.up.AREA above.timberline upland

(1.7) 49 ngga Ts’itae Tl’aa hwts’e’,

upland.PUNC river.flows.straight.headwaters (tpnm) AREA.to

‘Up above the treeline upland to ‘headwaters of river that flows straight’ (Sanford River headwaters),’

219


there.area.from DIST.upland.PUNC


bear.trap.river (tpnm)


there 1PL.SUB.pl.go.IMPERF.against.a.place

‘from there on upland we reached ‘bear trap creek’.’


00:01:45.390-00:01:59.290. Kari 2010:93))

Map 7.3: Area Described in (7.8)

Mr. Sanford begins this excerpt in IUs 46-47 by talking about going to the

uplands using a directional marked with the allative suffix -e(h) in ’unggeh

‘DIST.upland.ALL’. In the next IU he uses the area suffix to do the now familiar work

220

of laying out a major leg of his journey in the direction of the timberline (a different

region from the timberline in (7.7) above) using ’uttgu ‘DIST.up.AREA’, which he

then specifies in IU 49 as near the location called Ts’itae Tl’aa. Looking at Map 7.3,

we can see that the destination for this leg is quite far from its source at Natii Na’,

and that a number of named locations along the Sanford River intervene. In IUs 50-

52, Mr. Sanford names the first of these locations, Tsaani ’Ae Na’, using a

directional ’unggat ‘DIST.upland.PUNC’ marked with the punctual suffix. This is in

fact the first leg of a series of journey legs between Natii Na’ and the timberline

around Ts’itae Tl’aa. This major journey spans approximately 60 IUs and is

comprised of eight sequential minor legs. This leg, which represents a major episode

in the discourse (nearly half of Mr. Sanford’s entire narrative), ends with the excerpt

found in (7.9) below, in which Mr. Sanford arrives at the last named location, Dit’ox

Tl’aa, and moves into the canyon at Ts’itae Tl’aa (although Ts’itae Tl’aa is not

mentioned by name). The entire leg is shown in Map 7.4.

(7.9) End of major journey leg inititated in (7.8) 95 ((COUGH)) Dit’ox Tl’aa ts’e’,

3S.REFL.POSS.nest.headwaters to

(1.1) 96 Dit’ox Tl’aa ts’e’

3S.REFL.POSS.nest.headwaters to

kats’enaes you know. 1PL.SUB.move.nomadically.IMPERF.up

‘To ‘nest headwaters’, we would move to ‘nest headwaters’.’

97 Big glacier there you know. (0.3)

98 Dit’ox Tl’aa. 3S.REFL.POSS.nest.headwaters

221

(1.5) 99 Du’,

(0.2) 100 yii cu ts’edelts’iix debae gha,

3S.NH also, else 1PL.SUB.sit/stay.IMPERF sheep in.relation.to

‘We stayed there at ‘nest headwaters’ by the sheep.’

101 Sesyaann’, ram

102 sesyaann’ yaen’ una’ c’ilaen. ram only 3S.POSS.river be, be born

‘There are rams, only rams on that creek.’

(1.5) 103 Yii ts’eghaax xona.

3S.NH 1PL.SUB.kill.pl.O.IMPERF then

(0.3) 104 Yii kae yet ’sdalts’iix.

3S.NH INSTR there 1PL.SUB.sit/stay.IMPERF

‘We killed some and we stayed there on that.’

(1.1) 105 K’a xona yet hwts’en xona

EMPH then there from.AREA then

na’stetnaesi, 1PL.SUB.travel.nomadically.IMPERF.ITER.REL

(2.2) 106 ohh dahwtne dak.

AREA.SUB.be.steep.IMPERF

‘Then as we moved back from there, oh it (the canyon) was steep.’

(0.6) 107 Ni k’aedze’ dahwtne dak xona.

RECIP.cavity.ADVZ AREA.SUB.be.steep.IMPERF then

(0.6) 108 Saanetah kats’enaes.

barely 1PL.SUB.move.nomadically.IMPERF.up

‘It was steep on both sides, and then we could barely move up.’


00:03:53.300-00:04:06.320. Kari 2010:96))

222

Map 7.4: Route from Natii Na’ to Ts’itae Na’

In summary, (7.8) is the initiation of a long journey between Natii Na’ and

Ts’itae Tl’aa, in which Mr. Sanford names the end location of this major leg before

“backing up” and sequentially discussing all the salient named locations along the

way.

7.4 Reclaiming Distance Distinctions Via Discourse Use of

Suffixes

In addition to the allative connotation provided by the directional containing the

area suffix in setting the trajectory of the journey, note that the contrast between

locations marked with directionals containing the area suffix and locations marked

with directionals containing the punctual suffix is also serving a deictic function. We

223

have already seen that the distal prefixes are no longer fully productive in making a

clear distinction between near vs. far distances; the discourse use of the suffixes is

now taking over that semantic range in a sense. The suffixes are being used to make

distinctions in relative distance: the destinations of major journey legs are always

further away than those of the minor legs contained with them. As deictic prefixes

lexicalize and lose their semantic contrast, speakers are able to take advantage of

another part of directional grammar to serve the function of distinguishing distance.

The literal meaning of the area and punctual suffixes alone would not be enough to

predict that speakers would be able to repurpose the suffixes in this way. Instead, it

is how speakers use these suffixes in discourse to describe major and minor journey

legs and to create major and minor episodes in the discourse that allows speakers to

take advantage of deictic distinctions as well.

7.5 Conclusion: Grammaticalization in the Suffixes

This chapter and the previous have been concerned with the way directional

morphology is employed by speakers in discourse to serve the communicative

purpose of describing paths of travel through known locations in Ahtna territory. It

was shown that speakers’ metalinguistic knowledge of the morphology of the

directionals – while very sophisticated in an elicitation setting – is not equivalent to

speakers’ use of these words in spontaneous language production. This is not

surprising, however, because much research has shown (e.g. Bybee 2010, Hopper &

Traugott 2003)) that discourse fixedness leading to phonological fusion, semantic

bleaching, and lexicalization or grammaticalization is a common part of language

224

change. The directional prefixes, arising from demonstratives and historically

encoding a three-way deictic distinction, are an example of natural lexicalization.

Even within the narrow confines of a single travel narrative, in which the speaker’s

goal is to index his or her knowledge of cohesive swaths of Ahtna territory, speakers

do not consistently apply the directional prefixes in a way that is consistent with a

rich deictic function. The prefixes are fusing both phonologically and semantically to

the lexical directional stems, which is a natural result of their frequent occurrence in

spontaneous discourse.

The present chapter turned to the relationship between discourse and the

directional suffixes. It finds no indication of semantic bleaching, but discusses the

way the punctual and area suffixes are being used by speakers in discourse to not

only indicate literal points and areas, but to also segment the landscape into major

and minor legs, and to organize the discourse into a larger series of united smaller

episodes. In this way the area suffix also takes on an allative function in the

discourse structure, indicating to the listener the speaker’s eventual descriptive goal.

Furthermore, this segmenting of the journey into major legs that are made up of

multiple smaller legs allows speakers to convey relative distance in a way that the

prefixes no longer do.

In terms of a possible pathway for this kind of language change, consider Figure

7.2. Although we have no direct evidence of the historical semantics of the

directional affixes other than what has been posited in reconstructions like Leer

1989, we can assume for the sake of argument that the prefixes were once fully

deictic in that they encoded a three-way distinction in relative distance. We can

225

further assume that the suffixes were once used strictly for their grammatical

functions and did not serve any other purpose in structuring discourse. Through

discourse, however, natural processes of language change like lexicalization

occurred in the directional prefixes, stemming from relative frequency of use (in

particular of the highly frequent distal prefix ’u-). Furthermore, it is possible

discourse of a very specific genre, narration about travel and the spatial relationships

between referents in the physical geography, establishes a relationship between

temporal features, geographic features, event structure and narrative organization,

such that the suffixes now also accrue not only discourse-structuring functions, but

also deictic meaning.

Figure 7.2: Possible Pathway for Language Change Via Discourse

This is of course only a hypothesis, but it is based on a classic scenario of

lexicalization and grammaticalization, in which processes leading to loss of

semantic vividness are counterbalanced by processes that lead to a

reestablishment of vividness in another part of the grammar.

226

Chapter 8

Summary

This dissertation has been an examination of one grammatical system in Ahtna,

the lexical class of directionals. In particular, this dissertation has investigated

directionals from the point of view of language change on two fronts. First, the

semantics of the directional system was considered in light of the changes it is

undergoing as an endangered language in constant contact with the dominant English

language. My fieldwork with Mr. Pete, a fluent first-language speaker of Ahtna who

spends much of his time communicating in English, shows that the traditionally

riverine directional system is shifting to resemble the English cardinal directional

system. In particular, the principles of Major River Orientation are changing, and in

Mr. Pete’s use of the directionals, the major axis of the directional system no longer

consistently aligns with the major river in a given drainage. This was shown in his

reluctance to use the term nae’ ‘upriver’ in reference to the upriver direction in the

Matanuska drainage. Instead, he prefers to reserve nae’ exclusively for rivers that

flow from the north to the south. This is likely caused by the predominance of the

southerly-flowing Copper River in Mr. Pete’s home region, and by his daily need to

refer to the upriver direction in English as north. Since he can no longer use nae’ for

the eastern upriver direction of the Matanuska or other non-southerly-flowing rivers

(and indeed even for the section of the Copper that curves away from a north-south

trajectory), he must invoke other directionals to fill this gap in the paradigm. We

227

have seen that he selects one of two options. First, he uses ngge’ ‘upland’ to refer to

the upriver direction on the Matanuska, likely in reference to the upward sloping

valley in which the river flows. Second, he uses tgge’ ‘up (vertically)’, very possibly

because of contact with the semantic map of the English word up, which includes

non-vertical dimensions.

The second examination of language change herein was an investigation of the

morphology of the directionals as they are used in discourse. Using spontaneous

travel narrative data from a previous generation of fluent Ahtna speakers, it was

shown that despite an assumption in the literature that complex morphology of the

directional system leads to high token counts, in fact speakers restrict themselves to

a limited subset of the possible directional words, even when describing regions of

great physiographic diversity. This suggests that the processes of language change,

like phonological fusion and semantic bleaching are at play in the directionals.

Digging deeper, it was shown that the deictic prefixes, previously assumed to present

a distal vs. medial vs. proximal distinction (and possibly still retaining this

distinction in isolated situations like elicitation), are in fact becoming semantically

bleached in discourse. By comparing speakers’ uses of the deictic directional

prefixes with the actual distances they are describing in their travel narratives, it was

shown that speakers use the distal and proximal prefixes for a large range of possible

distances, and they use the medial prefix for only very small distances. The three-

way paradigm is somewhat obliterated in discourse, supporting the notion that

speakers do not have full access to the prior semantic range of the prefixes. The

228

prefixes are lexicalizing onto the directional stems and speakers are manipulating

these as single units.

Turning to the directional suffixes, it was shown that the area and punctual

suffixes are serving a function in the discourse above and beyond their literal

meaning. Speakers are systematically using a contrast between the two to do two

things. First, they are using the contrast to break the physical journeys through the

Ahtna countryside into major and minor legs. Locations that are named in the

narratives as long-range destinations are marked with the area suffix, and those that

are short range destinations on the way to the long-range destinations are marked

with the punctual suffix. This parallels the second use of the contrast between the

area and punctual suffix, which is to create major and minor episodes in the

discourse. The major episodes consist of the depicted events leading to the long-

range destinations, and the minor episodes are the depictions of events that occur

along the way. In this way, there is a relationship between the geographic use of the

suffixes, the temporal use of the suffixes, the events described and the overall

structure of both the discourse and the physical journey. Thus the suffixes, which

previously only encoded a distinction between a point and an area, are now also

accruing a deictic function via their use embedded in discourse: because the content

of the travel narratives is geography, and because there is a direct parallel between

the episodic structure of the discourse and the length of the minor and major journey

legs, the area suffix is now indicating locations that are far away, and the punctual

suffix is now indicating locations that are close. Discourse allows speakers to convey

distance in a way that the prefixes no longer reliably can.

229

Two other topics were investigated in this dissertation. The semantics of one of

the directional stems, nse’ ‘ahead’, is arguably difficult to pin down. Five possible

meanings of this term were investigated, none of which fully accounts for the data

collected during my fieldwork provided on its own. It seems that nse’ is best

described as a network of related semantic features, some arising from historical

meanings of the term, and others arising diachronically in the Ahtna territory. A few

semantic features seem to have developed into core senses of the term. These are

either ‘over a steep pass and to the edge of the ocean’, or ‘over a steep pass through

the Chugach’. Whichever of these two combinations is the synchronic semantic basis

of nse’, the salient feature here is the notion of a steep pass. Topography is salient in

the Ahtna directional system. Foot travel through mountainous areas requires a

different kind of effort from winter or summer travel on a waterway. The Ahtna

directional paradigm is more three-dimensional than the English system, which only

acknowledges differences in one plane.

Finally, the value of restricting descriptions of how Ahtna grammar encodes path

and location to data collected from a single genre of language use was examined. It

was shown that data limited to ‘frog stories,’ a common device in typological studies

of spatial concepts in language, do not fully reveal the richness of Ahtna spatial

grammar. Descriptions of this part (and by extension, most if not all parts) of the

grammar of Ahtna and other languages is better studied through spontaneously-

arising discourse in a genre that is indigenous to the speaker community.

230

Appendix

Transcription and glossing conventions

Transcription conventions

All transcriptions of spoken language are provided in the transcription system

outlined in Du Bois (2006) and Du Bois et. al. (1992), known as Discourse

Transcription. This system is based on the notion of the ‘intonation unit’, or IU, the

holistic prosodic unit that forms the armature upon which speakers build discourse

(see Berez to appear for more information on IUs in Ahtna). Each text line contains

one and only one IU, the boundaries of which were determined by a series of

acoustic and perceptual cues that are outlined in Berez (to appear).

The line breaks in the examples drawn from Kari (2010) do not match those

found in the original publication. Kari’s line breaks correspond to syntactic

sentences, but I have inserted line breaks at IU boundaries (with permission) for the

purposes of this dissertation.

Below are other symbols related to the transcription system that can be found in

the examples of spoken language herein:

17 IU number

(2.4) Pause length in seconds

MJ; Speaker identification label

, Continuing boundary tone

. Terminative boundary tone

231

? Appeal or rising boundary tone

@ Pulse of laughter

- Truncated word

_ Linked words

: Nonlexical segment lengthening

% Nonlexical glottal stop

# Inaudible syllable

As for free translations, in the examples drawn from my own fieldwork, I have

inserted free English translations of the Ahtna whenever enough Ahtna words

transpired between the interlocutors to warrant it for the sake of avoiding confusion.

Free translations in the examples drawn from Kari (2010) have been inserted at

locations in the transcript where my new IU-based line breaks align most closely

with Kari’s original line breaks. Note that I have not changed the wording of Kari’s

free translations, but I have inserted clarifications in square brackets where

necessary.

Glossing

All word-level glosses, except those that are otherwise cited, are my own. Kari

(2010) does not contain translations below the level of the sentence, so I have added

these to assist the reader. As with every other part of this dissertation, errors in

glossing are my own.

232

Morphemic parsing

The notorious fusional polysynthesis makes full morpheme-by-morpheme

parsing of every word cumbersome and furthermore this dissertation is not

concerned with fine-grained analyses of morphology of any part of the language

besides the directionals. Nonetheless, I have attempted to provide as much

information about the morphological makeup of words as possible without

unnecessarily cluttering the transcripts by committing to morpheme boundaries.

For most parts of speech, the morphological parsing is self-evident (e.g.

‘1PL.POSS.home’ for neghak’ae ‘our home’), but the picture is more complicated for

verbs. Rest assured, however, that there is a “method to the madness” in my glossing

of verbs, which does not strictly follow the order of the verb template from Kari

(1990). Verbs are glossed in the following order, for the categories that are present in

any given verb:

Information about the subject .

Information about the

object . Incorporated

stems .

Lexical meaning of verb .

Stem aspect .

Conjunct morphemes in L-R order .

Affixes (REL, NOM, ADVZ)

For example, a traditional fully-parsed gloss of the verb nits’enaes ‘we stopped’

would be as follows:

nits’- e- naes TERMIN-1PL.SUB- ASP-move.nomadically.IMPERF

Instead my glossing would be the following:

233

nits’enaes 1PL.SUB.move.nomadically.IMPERF.TERMIN

A more complicated example is na’stetnaesi ‘that we moved back’, which might

have a traditional parse and gloss as follows.

na- ’st- e- t- naes -i ITER-1PL.SUB- ASP- CLF- move.nomadically.IMPERF -REL

My glossing would be the following:

na’stetnaesi 1PL.SUB.travel.nomadically.IMPERF.ITER.REL

Glossing abbreviations

The following is a guide to abbreviations used in glossing.

1 first person

2 second person

3 third person

ABAN abandoned (e.g. word, IU)

ABL ablative

ADVZ adverbializer

ALL allative

AREA areal

CONJ conjunction

CUST customary

DEM demonstrative

DIST distal

234

EVID evidential

EXIST existential

FOC focus

FUT future

HES hesitation

HUM human

IMPERF imperfective

INCEP inceptive

INDEF indefinite

ITER iterative

LOC locative

MED medial

NEG negative

NH nonhuman

NOM nominalizer

O reference to object within the

lexical meaning of a verb

(from Kari 1990)

OBJ object

P reference to patient within the

lexical meaning of a verb

(from Kari 1990)

235

PERAMB perambulative

PERF perfective

PL plural

POSS possessor

PROX proximal

PUNC punctual

REFL reflexive

REL relativizer

S singular

SP/TMP spatial/temporal

SUB subject

TERMIN terminative

236

References

Allen, Gary L. 2000. Principles and practices for communicating route knowledge.

Applied Cognitive Psychology 14. 333-359.

Applebaum, Ayla Bozkurt & Andrea L. Berez. 2009. A theory is only as good as the

data: Casting a wide net in Kabardian and Ahtna documentation. In Peter K.

Austin, Oliver Bond, Monik Charette, David Nathan & Peter Sells (eds.),

Proceedings of Conference on Language Documentation and Linguistic

Theory 2, 29-38. London: School of Oriental and African Studies.

Ariane, Tom & Michel Denis. 2004. Language and spatial cognition: Comparing the

roles of landmarks and street names in route instructions. Applied Cognitive

Psychology 18. 1213-1320.

Axelrod, Melissa. 1993. The semantics of time: Aspectual categorization in Koyukon

Athabaskan. Lincoln: University of Nebraska Press.

Basso, Keith H. 1968. The Western Apache classificatory verb system: A formal

analysis. Southwestern Journal of Anthropology 24. 252-266.

Basso, Keith H. 1984. “Stalking with stories”: Names, places, and moral narratives

among the Western Apache. In Edward M. Bruner (ed.), Text, play, and

story: The construction and reconstruction of self and society, 19-55.

Prospect Heights, IL: Waveland Press.

Basso, Keith H. 1988. Speaking with names. In George M. Marcus (ed.), Rereading

cultural anthropology, 71-104. Durham: Duke University Press.

237

Basso, Keith H. 1996. Wisdom sits in places: Landscape and language among the

Western Apache. Albuquerque: University of New Mexico Press.

Bavin, Edith L. 1998. Factors of typology in language acquisition: Some examples

from Warlpiri. In Anna Siewierska & Jae Jung Song (eds.), Case, typology

and grammar: In honor of Barry J. Blake, 37-55. Amsterdam: John

Benjamins.

Bavin, Edith. 2000. Ellipsis in Warlpiri children’s narratives: An analysis of Frog

Stories. Linguistics 38(3). 569-89.

Bavin, Edith. 2004.Focusing on ‘where’: An analysis of Warlpiri Frog Stories. In

Sven Strömqvist & Ludo Verhoeven (eds.), Relating events in narrative,

Volume 2: Typological and contextual perspective, 17-36. Mahwah, New

Jersey; London: Lawrence Erlbaum Associates.

Bennett-Kastor, Tina. 2002. The “Frog Story” narratives of Irish-English bilinguals.

Bilingualism: Language and Cognition 5(2). 131-46.

Berez, Andrea L. To appear. Prosody as a genre-distinguishing feature in Ahtna: A

quantitative approach. Functions of Language.

Berez, Andrea L. 2009. Beyond Frog Stories: Speakers’ attention to path and

location in Ahtna discourse. Paper presented at the University of Melbourne

Department of Linguistics and Applied Linguistics, July 24, and the

University of Sydney Department of Linguistics.

Berman, Ruth A. & Dan I. Slobin.1994. Relating events in narrative: A

crosslinguistic study. Hillsdale, New Jersey; Hove, UK: Lawrence Erlbaum

Associates.

238

Boumans, Louis. 2006. The attributive possessive in Moroccan Arabic spoken by

young bilinguals in the Netherlands and their peers in Morocco.

Bilingualism: Language and Cognition 9(3). 213-31.

Brown, Penelope. 2004. Position and motion in Tzeltal Frog Stories: The acquisition

of narrative style. In Sven Strömqvist & Ludo Verhoeven (eds.), Relating

events in narrative, Volume 2: Typological and contextual perspective, 37-

58. Mahwah, New Jersey; London: Lawrence Erlbaum Associates.

Busch, John. 2000. Finding your way through a story: Direction terms in Gwich’in

narrative. University of Alaska Fairbanks MA thesis.

Bybee, Joan. 2010. Language, usage and cognition. Cambridge: Cambridge

University Press.

Chini, Marina & Ewa Lenart. 2008. Identifying the topic in a narrative task in Italian

and French by native speakers (L1) and learners (L2). Acquisition et

Interaction en Langue Étrangère 26. N.p.

Cook, Eung-Do. 1984. A Sarcee grammar. Vancouver: University of British

Columbia Press.

Cook, Eung-Do & Keren Rice (eds.). 1989. Athapaskan linguistics: Current

perspectives on a language family. Berlin: Mouton de Gruyter.

Cruikshank, Julie. 1990. Getting the words right: Perspectives on naming and places

in Athapaskan oral history. Arctic Anthropology 27(1): 52-65.

Denis, Michel. 1997. The description of routes: A cognitive approach to the

production of spatial discourse. Current Psychology of Cognition 16(4). 409-

458.

239

Denis, Michel, Francesca Pazzaglia, Cesare Cornoldi, & Laura Bertolo. 1999.

Spatial discourse and navigation: An analysis of route directions in the city of

Venice. Applied Cognitive Psychology 13. 145-174.

Du Bois, John W. 2006. Representing discourse. Ms., University of California, Santa

Barbara.

Du Bois, John W., Susanna Cumming, Stephan Schuetze-Coburn & Danae Paolino.

1992. Santa Barbara papers in linguistics, volume 4: Discourse

transcription. Santa Barbara: University of California, Santa Barbara

Department of Linguistics.

Duponceau, Peter Stephen. 1819. Report of the corresponding secretary to the

committee, of his progress in the investigation committed to him of the

general character and forms of the languages of the American Indians.

Transactions of the Historical and Literary Committee of the American

Philosophical Society, Vol. 1, xvii-xlvi. Philadelphia: American

Philosophical Society.

Engberg-Pedersen, Elisabeth & Frederikke Blytmann Trondhjem. 2004. Focus on

action in motion description: The case of West-Greenlandic. In Sven

Strömqvist & Ludo Verhoeven (eds.), Relating events in narrative, Volume

2: Typological and contextual perspective, 59-88. Mahwah, New Jersey;

London: Lawrence Erlbaum Associates.

Enrico, John. 2004. Toward Proto-Na-Dene. Anthropological Linguistics 46(3). 229-

302.

240

Galvan, Dennis & Sarah Taub. 2004. The encoding of motion events in American

Sign Language. In Sven Strömqvist & Ludo Verhoeven (eds.), Relating

events in narrative, Volume 2: Typological and contextual perspective, 191-

218. Mahwah, New Jersey; London: Lawrence Erlbaum Associates.

Hargus, Sharon. 1988. The lexical phonology of Sekani. New York: Garland.

Heine, Bernd & Tania Kuteva. 2005. Language contact and grammatical change.

Cambridge: Cambridge University Press.

Herman, David. 2001. Spatial reference in narrative domains. Text 21(4). 515-541.

Hoff-Ginsberg, Erika. 1997. Frog stories from four-year-olds: Individual differences

in the expression of referential and evaluative content. Journal of Narrative

and Life History 7. 223-27.

Hopper, Paul J. & Elizabeth Closs Traugott. 2003. Grammaticalization, 2nd edn.

Cambridge: Cambridge University Press.

Himmelmann, Nikolaus P. 1998. Documentary and descriptive linguistics.

Linguistics 36. 161-195.

Himmelmann, Nikolaus P. 2006. Language documentation: What is it good for? In

Jost Gippert, Nikolaus P. Himmelmann & Ulrike Mosel (eds.), Essentials of

language documentation, 1-30. Berlin: Mouton de Gruyter.

Holton, Gary. To appear. Differing conceptualizations of the same landscape: The

Athabaskan and Eskimo language boundary in Alaska. In David Mark,

Andrew Turk, Niclas Burenhult, & David Stea (eds.), Landscape in

language. Amsterdam: John Benjamins.

241

Holton, Gary. 2000. The phonology and morphology of the Tanacross Athabaskan

language. University of California, Santa Barbara dissertation.

Ibarretxe-Antuñano, Iraide. 2004a. Language typologies and our language use: The

case of Basque motion events in adult oral narratives. Cognitive Linguistics

15(3). 317-49.

Ibarretxe-Antuñano, Iraide. 2004b. Motion Events in Basque narratives. In Sven




Jetté, Jules & Eliza Jones. 2000. Koyukon Athabaskan dictionary. Fairbanks: Alaska

Native Language Center.

Jovanovic, Jelena & Aida Martinovic-Zic. 2004. Why manner matters: Contrasting

English and Serbo-Croatian typology in motion description. In Carol Lynn

Moder & Aida Martinovic-Zic (eds.), Discourse across languages and

cultures, 211-26. Amsterdam: John Benjamins.

Kang, Jennifer Yusun. 2004. Telling a coherent story in a foreign language: Analysis

of Korean EFL learners’ referential strategies in oral narrative discourse.

Journal of Pragmatics 36(11). 1975-90.

Kari, James. Forthcoming. A case study in Athabaskan geographic knowledge. In

David Mark, Andrew Turk, Niclas Burenhult, & David Stea (eds.),

Landscape in language. Amsterdam: John Benjamins.

Kari, James. 1979. Athabaskan verb theme categories: Ahtna. Fairbanks: Alaska


242

Kari, James. 1983. Ahtna place names list. Fairbanks: Copper River Native

Association and Alaska Native Language Center.

Kari, James. 1985. Can Native place names be preserved? Alaska Native News, July

1985. 8-12.

Kari, James (ed.). 1986. Tatl’ahwt’taenn Nenn’, The Headwaters People’s country,

Narratives of the Upper Ahtna Athabaskans. Fairbanks: Alaska Native

Language Center.

Kari, James. 1988. Local vs. regional place naming conventions in Alaskan

Athabaskan languages. In Sharon Hargus (ed.), Proceedings of the Twenty-

Third Western Conference on Linguistics, 233-249.

Kari, James. 1989. Some principles of Alaskan Athabaskan geographic knowledge.

In Mary R. Key and Henry M. Hoenigswald (eds.), General and Amerindian

ethnolinguistics, in remembrance of Stanley Newman, 129-150. Berlin:

Mouton de Gruyter.

Kari, James. 1990. Ahtna Athabaskan dictionary. Fairbanks: Alaska Native

Language Center.

Kari, James. 1996a. Names as signs: The distribution of ‘stream’ and ‘mountain’ in

Alaskan Athabaskan languages. In Eloise Jelinek, Sally Midgette, Keren

Rice, and Leslie Saxon (eds.), Athabaskan language studies: Essays in honor

of Robert W. Young, 443-475. Albuquerque: University of New Mexico

Press.

Kari, James. 1996b. A preliminary view of hydronymic districts in northern

Athabaskan prehistory. Names 44(4). 253-271.

243

Kari, James. 1997. Upper Tanana place names lists and maps. Ms., Wrangell-St.

Elias National Park.

Kari, James. 1999. Native place names mapping in Denali National Park and

Preserve. Ms., National Park Service.

Kari, James. 2006. Traditional cultural properties in the vicinity of Sparrevohn Long

Range Radar site. Ms., University of Alaska Anchorage.

Kari, James. 2007. Dena’ina topical dictionary. Fairbanks: Alaska Native Language

Center.

Kari, James. 2008. Ahtna place names list, 2nd edn. revised. Fairbanks: Alaska


Kari, James (ed.). 2010. Ahtna Athabaskan travel narratives: A demonstration of

shared geographic knowledge among Alaska Athabascans, told by Jim

McKinley, Frank Stickwan, Jake Tansy, Katie John, Adam Sanford.

Fairbanks: Alaska Native Language Center.

Kari, James & James A. Fall. 2003. Shem Pete’s Alaska: The territory of the Upper

Cook Inlet Dena’ina, 2nd edn. Fairbanks: University of Alaska Press.

Kari, James & Ben A. Potter (eds.). 2010. The Dene-Yeniseian connection

(Anthropological Papers of the University of Alaska New Series, Vol. 5 (1-

2)). Fairbanks: University of Alaska Department of Anthropology.

Kari, James & Siri Tuttle. 2005. Copper River Native places: Report of culturally

important places to Alaska Native tribes of southcentral Alaska. Glennallen,

AK: Bureau of Land Management.

244

Kelly, Klara & Harris Francis. 2005. Traditional Navajo maps and wayfinding.

American Indian Culture and Research Journal 29(2). 85-111.

Kingston, John. 2005. The phonetics of Athabaskan tonogenesis. In Sharon Hargus

and Keren Rice (eds.), Athabaskan prosody, 137-184. Amsterdam: John

Benjamins.

Krauss, Michael. 1964. Proto-Athabaskan-Eyak and the problem of Na-Dene: The

phonology. International Journal of American Linguistics 30(2). 49-83.

Krauss, Michael. 1965. Proto-Athabaskan-Eyak and the problem of Na-Dene: The

morphology. International Journal of American Linguistics 31(1). 18-28.

Krauss, Michael. 1973. Na-Dene. In Thomas A. Seobok (ed.), Linguistics in North

America, 903-78. The Hague: Mouton.

Krauss, Michael. 2005. Athabaskan tone. In Sharon Hargus & Keren Rice (eds.),

Athabaskan prosody, 55-136. Amsterdam, Philadelphia: John Benjamins.

Krauss, Michael E. & Victor A. Golla. 1981. Northern Athapaskan languages. In

June Helm (ed.), Handbook of North American Indians, Vol.. 6: Subarctic,

67-86. Washington: Smithsonian.

de Laguna, Frederica & Catharine McClellan. 1981. Ahtna. In June Helm (ed.),

Handbook of North American Indians, vol. 6: Subarctic, 641-64.

Washington: Smithsonian Institution.

Landau, Barbara & Ray Jackendoff. 1993. “What” and “where” in spatial language

and spatial cognition. Behavioral and Brain Sciences 16. 217-265.

Leer, Jeff. 1979. Proto-Athabaskan verb stem variation part one: Phonology.

Fairbanks: Alaska Native Language Center.

245

Leer, Jeff. 1989. Directional systems in Athapaskan and Na-Dene. In Cook, Eung-

Do and Keren D. Rice (eds.), Athapaskan linguistics: Current perspectives

on a language family, 575-622. Berlin: Mouton de Gruyter.

Levinson, Stephen C. 2003. Space in language and cognition: Explorations in

cognitive diversity. Cambridge: Cambridge University Press.

Lipka, Leonhard. 1990. An outline of English lexicology: Lexical structure, word

semantics, and word-formation. Tuebingen: Newmeyer.

Mayes, Patricia. 2003. Language, social structure, and culture: a genre analysis of

cooking classes in Japan and America. Amsterdam: Benjamins

McDonough, Joyce. 1999. On a bipartite model of the Athabaskan verb. In Theodore

B. Fernald & Paul R. Platero (eds.), The Athabaskan languages: Perspectives

on a Native American language family, 139-166. New York, Oxford: Oxford

University Press.

Meyer, Mercer. 1969. Frog, where are you? New York: Dial Books for Young

Readers.

Mithun, Marianne. 1999. The languages of Native North America. Cambridge:

Cambridge University Press.

Morford, Jill P. 2002. The expression of motion events in Homesign. Sign Language

& Linguistics 5(1). 55-71.

Moore, Patrick & Daniel Tlen. 2007. Indigenous linguistics and land claims: The

semiotic projection of Athabaskan directionals in Elijah Smith’s radio work.

Journal of Linguistic Anthropology 17(2). 266-86.

246

Morgan, Gary. 2002. Children’s encoding of simultaneity in British Sign Language.

Sign Language & Linguistics 5(2). 131-65.

Ordoñez, Claudia Lucia. 2004. EFL and native Spanish bilingual schools in

Colombia: A first look at bilingual adolescent Frog Stories. International

Journal of Bilingual Education and Bilingualism 7(5). 449-74.

Orsolini, Margherita, Franca Rossi & Clotilde Pontecorvo. 1996. Re-introduction of

Referents in Italian Children’s Narratives. Journal of Child Language 23(2).

465-86.

Payne, Thomas E. 1984. Locational relations in Yagua narrative. Work papers of the

Summer Institute of Linguistics, University of North Dakota 28. 157-192.

Reckord, Holly. 1979. A case study of Copper Center, Alaska (Alaska OCS

Socioeconomic Studies Program Technical Report Number 7). Anchorage:

Bureau of Land Management.

Reckord, Holly. 1983a. That’s the way we live: Subsistence in the Wrangell-St. Elias

National Park and Preserve (Anthropology and Historic Preservation

Cooperative Park Studies Unit Occasional Paper No. 34). Fairbanks:

University of Alaska.

Reckord, Holly. 1983b. Where raven stood: Cultural resources of the Ahtna region.

Fairbanks: University of Alaska Press.

Ragnarsdóttir, Hrafnhildur & Sven Strömqvist. 2004. Time, space and manner in

Swedish and Icelandic: Narrative construction in two closely related

languages. In Sven Strömqvist & Ludo Verhoeven (eds.), Relating events in

247

narrative, Volume 2: Typological and contextual perspective, 113-42.

Mahwah, New Jersey; London: Lawrence Erlbaum Associates.

Rice, Keren. 1989. A grammar of Slave. Berlin: Mouton de Gruyter.

Rice, Keren. 2000. Morpheme order and semantic scope: Word formation in the

Athapaskan verb. Cambridge: Cambridge University Press.

Sapir, Edward. 1915. The Na-Dene languages: A preliminary report. American

Anthropologist 17. 534-558.

Sapir, Edward. 1923. A type of Athabaskan relative. International Journal of

American Linguistics 2. 136-142.

Serratrice, Ludovica. 2007. Referential cohesion in the narratives of bilingual

English-Italian children and monolingual peers. Journal of Pragmatics 39(6).

1058-87.

Simeone, William E. & James Kari. 2002. Traditional knowledge and fishing

practices of the Ahtna of Copper River, Alaska (Division of Subsistence

Technical Paper No. 270). Anchorage: Alaska Department of Fish and

Game.

Simeone, William E. & James Kari. 2005. The harvest and use of non-salmon fish

species in the Copper River basin. Anchorage: Alaska Department of Fish

and Game.

Slobin, Dan I. 1996. Two ways to travel: Verbs of motion in English and Spanish. In

Matthew Shibatani & Sandra A. Thompson (eds.), Grammatical

constructions: Their form and meaning, 195-220. Oxford: Clarendon Press.

248

Slobin, Dan I. 2004. The many ways to search for a frog: Linguistic typology and the

expression of motion events. In Sven Strömqvist & Ludo Verhoeven (eds.),

Relating events in narrative, Volume 2: Typological and contextual

perspective, 219-58. Mahwah, New Jersey; London: Lawrence Erlbaum

Associates.

Strömqvist, Sven & Ludo Verhoeven. 2004. Relating events in narrative, Volume 2:

Typlogical and contextual perspectives. Mahwah, New Jersey; London:

Lawrence Erlbaum Associates.

Talmy, Leonard. 1985. Lexicalization patterns: Semantic structure in lexical form. In

Timothy Shopen (ed.), Language typology and lexical description, Volume 3:

Grammatical categories and the lexicon, 36-149. Cambridge: Cambridge

University Press.

Talmy, Leonard. 1991. Path to realization: A typology of event conflation.

Proceedings of the Berkeley Linguistics Society 17. 480-519.

Talmy, Leonard. 2000. Toward a cognitive semantics, Volume II: Typology and

process in concept structuring. Cambridge, Massachusetts: MIT Press.

Tenenbaum, Joan. 1978. Morphology and semantics of the Tanaina verb. Columbia

University dissertation.

Thomason, Sarah Grey & Terrence Kaufman. 1988. Language contact, Creolization,

and genetic linguistics. Berkeley: University of California Press.

Thornton, Thomas F. 2008. Being and place among the Tlingit. Seattle, London:

University of Washington Press in association with Sealaska Heritage

Institute, Juneau.

249

Verhoeven, Ludo. 2004. Bilingualism and narrative construction. In Sven Strömqvist

& Ludo Verhoeven (eds.), Relating events in narrative, Volume 2:

Typological and contextual perspective, 435-54. Mahwah, New Jersey;


Wilkins, David P. 2004. The verbalization of motion events in Arrernte. In Sven




Winskel, Heather. 2007. The expression of temporal relations in Thai children’s

narratives. First Language 2. 133-58.

Young, Robert W. & William Morgan. 1980. The Navajo language: A grammar and

colloquial dictionary. Albuquerque: University of New Mexico Press.

Zlatev, Jordan & Peerapat Yangklang. 2004. A third way to travel: The place of Thai

in motion event typology. In Sven Strömqvist & Ludo Verhoeven (eds.),

Relating events in narrative, Volume 2: Typological and contextual

perspective, 159-90. Mahwah, New Jersey; London: Lawrence Erlbaum

Associates.

Directional reference, discourse, and landscape in Ahtna

Documents

Transcript of Directional reference, discourse, and landscape in Ahtna