The Reflexive Relativism of Georg Simmel

The Reflexive Relativism of Georg Simmel

Jared A. Millson

The Journal of Speculative Philosophy, New Series, Volume 23, Number3, 2009, pp. 180-207 (Article)

Published by Penn State University PressDOI: 10.1353/jsp.0.0077

For additional information about this article

Access provided by Agnes Scott College (29 Oct 2013 10:57 GMT)

http://muse.jhu.edu/journals/jsp/summary/v023/23.3.millson.html

http://muse.jhu.edu/journals/jsp/summary/v023/23.3.millson.html

jsp The Refl exive Relativism of Georg Simmel

Jared A. Millson emory university

The term relativism covers a variety of different philosophical views. This diversity may be organized and understood by highlighting an underlining structural similarity. Nearly all forms of relativism are distinguishable by what is relative (e.g., truth, meaning, rationality, ethics, etc.) and what these things are relative to (e.g., cultures, individuals, epochs, etc.). Those items that fi t in the former category may be termed the “topic,” and those in the latter may be called the “perspective.” While these terms are only placeholders, my choice of the word perspective refl ects the general belief shared by relativ-ists that certain empirical and existential circumstances endow individuals with a unique point of view on the world. In order to do justice to this par-ticularity, relativists insist that we must treat the occupant(s) of a “perspec-tive” as the ultimate arbitrator(s) of its normative categories.

Among the oldest and most common objections raised against relativ-ism is one known as the self-refutation argument. First developed by Plato against Protagorean relativism in the Theaetetus , this argument involves the application of relativism to itself. 1 It is argued that in order to assert the rela-tivity of truth, knowledge, or “everything,” the validity, warrant, truth, and so forth of this claim itself either must be absolute, in which case it presup-poses the absolute truth it claims to deny, or must be relative. If it is relative, then according to some other “perspective” the claims of absolutism are,

journal of speculative philosophy, vol. 23, no. 3, 2009 Copyright © 2010 The Pennsylvania State University, University Park, PA

JSP_23-3_02.indd 180JSP_23-3_02.indd 180 12/18/09 10:21 AM12/18/09 10:21 AM

the reflexive relativism of georg simmel 181

or could be, true. However, it is the nature of absolutism to be true across all perspectives; thus, on pain of contradiction, relativism must be false in order for absolutism to be true. Most responses to this dilemma take one of three options. One response is to accept the fi rst horn of the dilemma and to mitigate the claims of relativism by acknowledging that at least one claim is true, justifi ed, and so on absolutely, namely, the claim of relativism. Another response takes the second horn and restricts the validity of relativism to the “perspective” of the relativist while denying the possibility of absolutism’s nonperspectival truth. These two options— absolute relativism and refl exive relativism —are joined by a third that obviates the choice between alterna-tives by denying the dilemma itself. Someone offering this response might suggest that the criticism begs the question by assuming the very criteria of truth and acceptability that the relativist puts in question. The existence of absolute criteria, upon which the absolutist bases the self-refutation argu-ment, is exactly the point at issue, so they cannot be used to fault relativism.

Without assessing the viability of any of these abstract, hypothetical responses, I would like to explore a particular formulation of refl exive rel-ativism that indicates a path leading from refl exivity to the global denial represented by the third response to the charge of self-refutation. It is my contention that the relativism articulated by Georg Simmel in The Philoso-phy of Money pursues the consequences of refl exivity up to point at which the very opposition between accepting and rejecting absolutism dissolves. Although he is rarely read as a contributor to debates in epistemology, Simmel may be seen as a modern philosopher who believed in the insep-arability of philosophy (including epistemology as one of its fi elds) and sociology. In The Philosophy of Money , Simmel defends an epistemological position that refl ects, supports, and complements the methodology with which he analyzes social phenomena. 2 While I recognize that my focus on the epistemological issues raised in Simmel’s work to the exclusion of his methodological and sociological concerns amounts to an artifi cial separa-tion, I believe that doing so will allow us to bring to light Simmel’s relevance for problems raised in contemporary epistemology.

I call Simmel’s an epistemological refl exive relativism because its claims involve the nature of truth and knowledge. By employing this broad notion of epistemology I am not seeking to avoid the diffi cult task of distinguishing between knowledge and truth. However, as will hopefully become clear, Sim-mel is committed to an epistemic theory of truth. That is to say, he believes that truth is intelligible only in terms of epistemic categories like belief, assertion,


jared a. millson182

and inference and is best understood as a functional relationship that develops among these elements in the course of inquiry. Conversely, we fail to grasp the meaning of truth when we try to isolate it from our processes of verifi cation. This is not a position that Simmel argues for explicitly. Rather, it represents an implicit commitment (one that Simmel may not have fully understood him-self) articulated alongside his response to the issue of relativism. In an effort to highlight the various epistemological claims developed in Simmel’s treatment of relativism, I have chosen to employ terms of twentieth-century epistemology (e.g., foundationalism, coherentism, pragmatism , etc.) anachronistically in my exposition. It is my hope that doing so will enable us to observe the relevance of Simmel’s relativism for contemporary epistemology as well as lay the ground for a better understanding of the relationship between Simmel’s epistemology and his sociological methodology.

Simmel begins his discussion of relativist epistemology with the claim that “the fi rst tendency of thought” is to focus attention on what is “stead-fast and reliable behind ephemeral appearances and the fl ux of events” (102; 94). Unable to reconcile themselves with change, “early modes of thought” pursue a search for absolute, fi xed points of support for intellec-tual activity (102; 94). While “almost all cultures” originally adopted this approach to fundamental questions of truth, knowledge, and meaning, “the basic tendency of modern science is no longer to comprehend phe-nomena through or as specifi c substances, but as motions, the bearers of which are divested of any specifi c quality. . . . [I]t [modern science] has abandoned the search for the essence of things and is reconciled to stat-ing the relationships that exist between objects and the human mind from the viewpoint of the human mind” (103; 95). As might be expected, Kant’s Copernican Revolution serves Simmel as the paradigmatic instance of this transformation. He holds that the “dissolution of absolute objectivity” via the grounding of the objects of knowledge in the conditions for the possibility of cognition represents a “psychological derivation” that never-theless depends upon “axioms which cannot have a merely psychological signifi cance” (103; 95–96). The claim that Simmel is attempting to articu-late here is that by locating the basis of knowledge in the principles neces-sary for human cognition, Kant found a new “absolute” in the structures of the human mind. In this way, Kant’s epistemological scheme, “even if carried to its conclusion, would still allow, or even require, a fi xed point, an absolute truth” (103; 95). What Simmel wants to show, in contrast to Kant,



is that if the relativism implied by modern science is taken seriously, then it requires us to treat a priori structures of consciousness in general as deter-minant of the empirical realm while allowing for the empirical and histori-cal constitution of particular a priori norms themselves. This relationship of reciprocal dependence between a posteriori and a priori propositions is summed up in the following manner: “If on the one hand we have the task of seeking in every present phenomena . . . the persistent a priori norms by which it is formed; then on the other, stands the maxim that we should attempt to trace every individual a priori (but not the a priori as such) back to its genesis in experience” (114, translation altered; 112). As we shall see, the notion of reciprocity ( Wechselwirkung ) will become crucial to Simmel’s treatment of relativism.

Before pursuing this treatment, however, Simmel needs to explain how his take on the issues will relate to this historical account. Though he admits that his own epistemology is historically situated as the successor to the “early modes of (absolutist) thought,” he nevertheless understands that in order to remain consistent, this epistemology must not rely on an uninterro-gated historical narrative. Indeed, the question of what is to count as histori-cal knowledge is itself raised by the emergence of modern relativism and its fascination with the perspectival constitution of knowledge claims. The tran-sition of premodern to modern modes of thought is not written as an insular logical development of the mind or as a change in arational social relations that “determine” the forms of consciousness; rather, the movement between absolute and relative thought forms is itself an issue internal to the concerns and claims of Simmel’s relativism. The knowledge and truth of the historical passage between worldviews requires further relativization. Therefore, Sim-mel’s epistemology, though self-conscious of its historicity, does not rely on the absolute validity of an external historical narrative in order to achieve its refl exivity.

It is my contention that Simmel formulates his relativist epistemology by outlining a series of four viewpoints, each of which reveals a dimension or aspect of his relativism. Aside from representing a brilliant integration of expositional form and theoretical content, this method of presentation enables Simmel to play off the paradoxes of unity and multiplicity that emerge in any discussion of relativism. His different formulations emphasize aspects that are not obviously symmetrical and which at times appear contradictory or at least incongruent. In the rest of the article, I will analyze the structure


jared a. millson184

and implications of what I am calling Simmel’s refl exive relativism as it is articulated in The Philosophy of Money . While it is true that each of the four formulations depends on the others for clarifi cation, elaboration, and com-pletion, they are not all equally signifi cant. The fi nal formulation, heuristics, occupies a privileged position in Simmel’s exposition because it resolves ambiguities and tensions that appear in the fi rst three. It is for this reason that we must follow the trajectory of his own account in the course of our own analysis.

Fallibilist Foundationalism

The fi rst formulation of Simmel’s epistemological position appears as a defense of fallibilism with regard to foundational knowledge. He argues that “the truth of any proposition can be known only on the basis of criteria that are completely certain and general” (103; 96). However, if these criteria are to have the status of certainty, they must be “legitimated by higher-level cri-teria, in such a fashion that a hierarchical series of cognitions is constructed, the validity of each one depending upon the preceding one” (103; 96). If the whole series is not to be “suspended in the air,” then an ultimate foundation must be sought from which all other propositions may be derived but which itself needs no such legitimation.

Although he never mentions his debt to the ancient skeptics, what Sim-mel has laid out here is essentially the Dilemma of the Criterion designed by Sextus Empiricus to generate skepticism and encourage the suspension of judgment. 3 The argument is relatively straightforward: Demonstrating the truth of any claim requires a criterion of truth with which to judge it. A truth criterion determines a claim’s truth when the latter can be inferred from the former, leaving the precise character of inference (i.e., logical, epistemic, material, etc.) unspecifi ed. As soon as one introduces such a criterion, how-ever, it too becomes a claim requiring another criterion of truth to determine its truth. The result, according to Sextus, is either an infi nitely regressive inferential chain or an admission of circular reasoning, that is, that at some point our deployment of criteria brings us back to our original claim. As we shall see, Simmel takes both results as crucial pieces in his account of relativ-ism. But for the moment it is worth noting that his discussion ends with a description of foundationalism.



Foundationalism in this context, and perhaps in its most general for-mulation, is a response to the Dilemma of the Criterion that takes the fol-lowing lines: If we are to determine the truth of any particular claim, and if doing so means testing whether it concludes a correct inference from propositions whose truth is likewise inferentially known, then there must be some criteria or propositions that are known noninferentially and which, therefore, do not require further support. All possible propositions can thus be assessed by relating them back to these basic items of direct, foun-dational knowledge. Different forms of foundationalism give different accounts of what it means for propositions to be known directly . Generally they are taken to be noninferential or self-evident truths. This is an episte-mological characterization because it deals with the kind of knowledge we have of such truths, but it is important to note that the claim often comes merged implicitly or explicitly with a metaphysical characterization of the truths in question as necessary . Sometimes the claim is interpreted to refer to self-justifying beliefs (i.e., beliefs the mere holding of which guarantees their truth); other times it means they are described as intuitive, indubita-ble, indefeasible, or a priori cognitions, although the last of these adjectives is not always coextensive with epistemically direct cognitions. The particu-lar kinds of basic criteria chosen by different foundationalist schemes are not what interest us here; for the moment what does is the fact that all such schemes are essentially theories of knowledge and/or justifi cation . Much like the distinction between truth and knowledge, the contrast between justifi -cation and knowledge is important. Under the classical theory, justifi cation, truth, and belief are considered to be the separately necessary and jointly suffi cient conditions of knowledge. Unfortunately, Simmel never treats the issue of justifi cation independently of knowledge. For this reason, I will take foundationalism strictly as a claim about knowledge that may or may not apply to justifi cation. Moreover, since he frequently talks about the pro-cess of arriving at true propositions, I will use the term verifi cation to refer to the epistemic process or processes by which we come to have knowledge, that is, how we come to possess true propositions.

Simmel’s approach to the foundationalist response to the Dilemma of the Criterion illuminates the fi rst major aspect of his relativism. He con-tends that while the “formal existence” (i.e., the possibility) of absolute, foundational knowledge may be established, its “real content” can never be determined, since “the attempt to fi nd an antecedent for what has been


jared a. millson186

taken to be the ultimate principle, is endless” (103–4, translation altered; 96). He notes,

No matter what proposition we have discovered as the ultimate one, standing over the conditionality of all other propositions, it remains possible that we shall recognize this one too as being merely rela-tive and conditioned by a superior one. This possibility is a positive challenge, which the history of thought has illustrated many times. Somewhere knowledge may certainly have an absolute basis, but we can never state irrevocably where this basis is; consequently, in order to prevent thought from arriving at dogmatic conclusions, we ought to treat each position at which we arrive as if it were the penultimate one. (104, translation altered; 96)

Thus far, Simmel appears to be formulating the following principle: since our experience of investigating knowledge claims demonstrates that every proposition that we have previously taken to be the ultimate basis of knowl-edge has been shown to be dependent upon yet “higher” propositions, we ought to treat every newly proposed ultimate truth as if it were itself condi-tioned. This regulation stems from the meta-inductive claim that because past assurances regarding a proposition’s self-evident or noninferential character have later turned out to be mistaken—that is, the proposition turns out to be inferable from other, “higher” propositions—present and future claims to foundational knowledge must be treated provisionally, that is, as potentially indirect and inferential.

This meta-inductive claim is characteristic of fallibilism, the position that all knowledge falls short of absolute certainty, that knowledge claims must remain open to the possibility of future revision, and that, at best, our claims can approximate indubitability. In its unqualifi ed and universal application, fallibilism is inherently opposed to the notion that knowledge can or ought to be founded on necessarily true propositions. Simmel’s prin-ciple sounds a lot like this defi nition of fallibilism, but it is not identical to it. Instead of subscribing to fallibilism simpliciter, he exploits the ambiguous status of this claim as both epistemological and metaphysical to develop a fallibilist foundationalism. Simmel’s own claim is that there may in fact be necessarily true or “absolutely” true propositions. If these exist—and the fact that “existence” is what is at stake here indicates the metaphysical qual-ity of this claim—they would certainly form the ideal foundations of our



knowledge. Unfortunately, the epistemological process by which we come to know such truths, that is, our verifi cation process, is essentially infer-ential since, following the scheme of the Criterion Dilemma, true propo-sitions are only known on the basis of other true propositions. Thus at any point that we have what appear to be self-evident propositions that are necessarily true independent of their relationship with other propositions, we are unable to verify them. At best, self-evidence means self-evident now, for us , and is taken as an indication of a claim’s provisional foundational status, that is, it operates as a grounding knowledge claim until it can be shown to derive from yet “higher” propositions. Simmel is not denying but, rather, affi rming the general possibility that there exists a self-validating, nec-essarily true proposition that could ground all knowledge claims. Indeed he avers that the notion of a system of knowledge founded upon an uncondi-tional truth represents the ideal that “has to be assumed” as the unreach-able goal toward which our concrete inquiries strive (104–5; 97–98). He qualifi es this general possibility with the assertion that all particular judg-ments about the foundational status of specifi c propositions are susceptible to revision. It is not that the foundational principles may themselves be false but, rather, that they may not really be epistemically direct, noninfer-ential, or self-validating such that they are necessarily true, and hence our claims about them (i.e., about them being absolute) could be false. He is thus not proposing a general theory of fallibilism that would deny the pos-sibility of foundational knowledge in general, but only a fallibilism with regard to determinant judgments about the content of that knowledge.

Simmel illustrates this point by noticing that the juridical system, in which every law is legitimated by another, ultimately rests on a “self-sub-sistent right [ in sich selbst ruhende Recht ]” that cannot be established legally. The power that represents this extralegal foundation “strives for its legiti-mation or for the fi ction of legitimacy, as if in homage to that absolute right, which lies beyond all relativity and is never captured by it, but which, for us, fi nds its symbolic representation in the continuous deduction of every exist-ing legal rule from a preceding one” (105, translation altered; 98–99). Sim-mel’s twofold contention is that the relational verifi cation that characterizes our knowledge continues to evoke the idea of an absolute starting point for knowledge and yet this relationality also suggests that any candidate we offer for this foundational truth has the possibility of being conditioned by unknown presuppositions. He notes the provisional character of our “abso-lute principles” in the following manner: “Our knowledge rests upon fi rst


jared a. millson188

principles which cannot be verifi ed at any given moment, because without these we should not arrive at the relative series of derived proofs; but they do not possess the logical dignity of being verifi ed. They are not true in the same sense as that which is verifi ed, and our thinking holds on to them as ultimate points only until it is able to reach beyond them to a higher level at which that which was hitherto accepted as axiomatic can be verifi ed ” (105, translation altered, emphasis added; 99). I emphasize the last portion of this passage because it indicates that, according to Simmel, the “fi rst prin-ciples” from which all verifi cation proceeds, our foundational knowledge, are to be taken provisionally , retaining their grounding epistemic role yet remaining open to possible revision. Again, stressing the provisionality of knowledge is a hallmark of fallibilism in general, and so it is not surprising to see Simmel making this point. What is surprising is that provisional-ity is attributed to knowledge claims regarding what ought to be epistemi-cally direct, infallible, foundational principles. His point seems to be that we can only verify our truth claims on the basis of other items of knowl-edge, which, despite functioning as criteria of verifi cation, are not thereby immune to falsifi cation. The idea of arriving at a fi xed, absolute, and inde-feasible principle is not theoretically impossible. But at any particular point in the regress a presupposed item of knowledge only functions foundation-ally vis-à-vis the particular claim in question. In other words, the question of a principle’s foundational function is always a matter of which claims it is foundational for , that is, where the regressive epistemic chain began. It is not thereby contradictory to imagine such a principle losing this grounding function and being itself grounded on yet higher principles when the par-ticular grounded claim in question is different. Understanding epistemic foundation in functional terms sheds light on Simmel’s claim that certain propositions or beliefs can both ground knowledge and be provisional. It also intimates the way in which verifi cation is a process that hinges on the inferential relations among items of knowledge.

Epistemic Circularity

Simmel’s second formulation of his relativist epistemology appears from the perspective of a position that I will call epistemic circularity. If, as Sim-mel has just argued, all particular claims to know the foundation of our knowledge are susceptible to revision, then it would seem that our process



of verifi cation inevitably leads back to the very infi nite regress to which foundationalism served as a response. Even if the discovery of an uncon-ditioned truth is the ideal intimated in this process, Simmel nevertheless describes the latter as “endless” [ niemals an seinem Ende anlangen kann ]. He rejects the foundationalist argument that the end to such a regress must be sought in a determinate unconditioned truth, since his fallibilism suggests that any determination of foundational knowledge is always open to revi-sion. He likewise rebuffs the skeptical conclusion that an infi nite regress necessarily implies that knowledge is impossible because it is premised on the foundationalist assumption that the series must have a self-vali-dating and absolute starting point. From the elimination of these alterna-tives, Simmel forges a new approach to the problem that views the infi nite regress as a positive condition for knowledge:

If one considers the vast number of hierarchically ordered presuppo-sitions, stretching to infi nity, upon which all knowledge with determi-nate content depends, there appears to be absolutely nothing barring the possibility that statement A is verifi ed by statement B, and state-ment B by the truth of C, D, E, etc., until fi nally it can only be veri-fi ed by the truth of A. The chain of reasoning C, D, E, etc. needs only to be suffi ciently long so that the return to the starting point eludes consciousness. . . . The interrelationship that we assume in our knowledge of the world—that from every point we can attain by demonstration every other point—seems to make this plausible. If we do not want to hold dogmatically once and for all to a single truth that permits of no verifi cation, it is easy to assume that this reciprocity of mutual verifi cations is the basic form of knowledge, conceived in its perfect state. Cognition is thus a free-fl oating process, whose elements determine their position reciprocally. (106, transla-tion altered; 99–100)

What we see here is Simmel taking up the second outcome of the Dilemma of the Criterion foreseen by Sextus: circularity . Interestingly, Simmel fi nds no problem with this situation. Instead, he thinks that the possibility of return-ing to the original claim in the regress of criteria invoked to verify it is indic-ative of the very structure of our knowledge. This view represents a variant of what has been called coherence theories of truth. 4 Such theories are epis-temic in the sense that truth is a particular (often inferential) arrangement


jared a. millson190

among propositions, beliefs, cognitions, and so on. In particular, this view is what I would call that of epistemic circularity because it refers to a structure of verifying principles that forms a circle, rather than a cluster, a constella-tion, or some other spatial model for the confi guration of epistemic support.

One problem that arises from this idea of knowing as a chain of rea-soning “only to be suffi ciently long so that the return to the starting point eludes consciousness” is that it sets up a model for knowledge that bears little resemblance to our ordinary, fi nite epistemic activities. This is why Simmel qualifi es his remarks by suggesting that such a theory of knowl-edge is one whose possibility is not barred, one that might represent the “basic form of knowledge, conceived in its perfect state ” (106; 100). In other words, he is presenting a response to the Dilemma of the Criterion that avoids skepticism by projecting an ideal state of knowledge indicated by our particular inquiries. He is not ignorant of the fact that the circular character of knowledge stands in contrast to our concrete and particular investiga-tions. Indeed, he notes that

the vast majority of our representations are taken for granted and the question of truth is usually applied only to a particular case. A judg-ment is then made in terms of the harmony or contradiction of this instance with the total complex of representations that are available and assumed to have already been established. On another occasion, any representation in the whole complex may become questionable, and the one to be investigated may belong to the determinant major-ity. The tremendous quantitative disproportion between the number of representations that are questionable and those that are secure and validated also helps to conceal the relation of reciprocity. In the same way, the disproportion of weights caused us, for a long time, to observe the gravitational attraction of the earth upon the apple but not that of the apple upon the earth. (115, translation altered; 115)

Simmel makes clear that our concrete epistemic activities of verifying prop-ositions always occur against a background of propositions and criteria that are not amenable, within the context of the particular inquiry, to verifi ca-tion. The truth of any particular proposition will be determined by its fi t with other items of knowledge functioning normatively as verifi cation cri-teria. These other items are not arbitrary propositions but presuppositions epistemically relevant to the particular proposition in question, that is, they



are inferentially related. (Since Simmel never even addresses the precise character of this relation or what kind of inference is involved, I will not get into the thorny business of speculating on how this relation of “relevance” is to be interpreted.) Just as we feel no problem measuring the weight of the apple as a function of the earth’s gravitational pull on it while ignoring the apple’s or any other object’s gravitational effects on the earth, so it is with truth. We certainly recognize the possibility of testing the validity of every relevant presupposition involved in the verifi cation of a particular proposi-tion, much as we recognize the possibility of measuring the gravitational pull exerted by the apple—or indeed any physical object in the universe—upon the earth, but we would never consider it necessary to carry out these measurements in order to ascertain the weight of the apple.

So we fi nd ourselves in need of verifying certain propositions by assess-ing their relation with other relevant propositions whose truth is assumed, for the purpose of concrete inquiry. The possibility of placing every pre-supposition and its presuppositions under examination is never denied, but much like the attempt to measure the reciprocal gravitational effects of every object in the universe, it is deemed an infi nite task that remains only an ideal possibility. The circularity of knowledge is likewise an ideal pos-sibility that our concrete inquiries suggest but which is never actualized. In other words, while the contextual limitations of our acts of verifi cation indicate the possibility of an infi nite circularity to our knowledge, this pos-sibility in no way delegitimates our actual inquires because of their fi nitude. Finally, Simmel’s example suggests that the fi nitude of our concrete inqui-ries is not just a function of contextual limitations imposed by the sub-ject matter itself but is, rather, part of the inherently purposive character of those inquiries. Our particular verifi cations are always determined, in part at least, by the reasons why we undertake inquiry in the fi rst place. Thus, while it is indeed possible to measure the infi nitesimal pull of the apple on the earth, there is virtually no reason to do so. (We will want to keep this purposive quality in mind as we turn to Simmel’s pragmatic formulation of relativism.)

The relation between this ideal epistemic circularity and Simmel’s fal-libilism with regard to foundational knowledge is now emerging. To say that certain items of knowledge only function foundationally in relation to the particular claims under examination and that such items could very well turn out to be themselves conditioned yields the same consequences as the assertion that particular claims are verifi ed against the backdrop of


jared a. millson192

items of knowledge whose own verifi cation has not yet been performed. Moreover, there is little substantive difference between (1) the notion that it is possible that all knowledge rests on absolute, unconditioned items of knowledge but that the determination of its content (i.e., what the foun-dational principles actually are) involves an infi nitely regressive inferential chain; and (2) the idea that items of knowledge are linked inferentially into a massive circle whose transversal is only an “ideal possibility.” In both cases, the contextual and inferential character of concrete inquiry projects an ideal structure of knowledge whose confi rmation is both outside the capacities of our fi nite minds and beyond the interests of embodied subjects. The opposition between an infi nite regress and a massive circle of reasoning is obviated by the fact that both are the ideals intimated in and by our concrete epistemic activities. Thus, Simmel’s primary response to the Dilemma of the Criterion is to suggest that the foundational and circular responses are equally plausible and entirely compatible so long as they are seen for what they are: the ideally possible structure of knowledge that is projected by, yet unverifi ed in, our concrete, mundane, and fi nite inquiries. The latter are always carried out in reference or response to a particular problematic claim and against the backdrop of cognitions whose truth is taken provisionally.

Pragmatic Naturalism

One objection that has not been raised thus far is that even Simmel’s depic-tion of verifi cation in these mundane inquiries is dependent upon proposi-tions that are already within the sphere of knowledge even if they are only provisional norms. The question thus arises: How do propositions originally enter the system of knowledge? Simmel’s third formulation of relativism, what I will term pragmatic naturalism, takes up this question. He begins with the position of what might be described as a biologized Kantianism: “All representations of entities are functions of a specifi c physical and psy-chological organization which does not mirror those entities in any mechan-ical way” (106; 100). Simmel steers clear of the skepticism this assertion might entail by arguing that our representations nevertheless “form the presuppositions, the material and the directives for our practical activity, through which we establish a relationship with the world as it exists in rela-tive independence of our subjectively determined representation” (106; 101). Truth thus lies in the relationship that holds between our representations



and our practical engagement with the world. Simmel describes this rela-tionship in the following passage:

It is a very striking fact that actions carried out on the basis of rep-resentations that are not at all identical with objective being never-theless secure results of a reliability, expediency and accuracy that could hardly be greater if we knew the objective conditions as they are in themselves, whereas other activities based on “false” repre-sentations end in real injure to us. . . . Whether an action guided and determined by a representation will have useful consequences for the agent cannot be determined by the content of this represen-tation, even though it might correspond with absolute objectivity. The result will depend entirely upon what this representation can accomplish as a real process within the organism, in cooperation with other physical and psychological forces and with reference to the specifi c needs of life. If we assert that man sustains and supports life only on the basis of true representations, and destroys it by false ones, what does this “truth”—the content of which is different for each species and which never refl ects the true object—mean except that some representation associated with particular organization and its powers and needs leads to useful results? Initially, truth is not useful because it is true but vice-versa. We dignify with the name “truth” those representations that, active within us as real forces or motions, incite us to useful behavior. Thus there are as many basi-cally different truths as there are different organizations and condi-tions of life. (107; 102)

Many of these claims resonate with the pragmatism of William James and other pragmatists for whom “an idea is ‘true’ so long as . . . it is profi t-able to our lives.” 5 Truth, for Simmel, consists in the usefulness of a repre-sentation (proposition, judgment, etc.), insofar as it serves the satisfaction of “the specifi c needs of life.” All the elements of a correspondence theory of truth are present—subjectively determined representations and an objec-tive reality—and yet Simmel insists that any “correspondence” between our concepts and their objects is strictly irrelevant to the issue of truth. Propositions are true only insofar as their practical employment leads to consequences that maintain or improve the life of the organism. Moreover, since each species has its own particular physiopsychological constitution


jared a. millson194

(i.e., representative faculty) and its own set of vital needs and powers, each species will have its own truth. This addresses one of the primary ques-tions that is put to the pragmatist: If truth consists in the utility of a con-cept to the achievement of certain purposes, then isn’t truth relative to the purposes arbitrarily chosen by the believer? Simmel’s response appears to lie in the claim that the “purposes” in question are those needs particular to each species yet necessary for their individual members. Insofar as the ends for which true propositions are useful derive from the natural orga-nization of beings, I feel justifi ed in describing Simmel’s theory of truth as pragmatic-naturalism (which is of course not to be confused with other meanings of naturalism that refer to attempts to wed epistemology to the methods and discoveries of natural science).

It appears that Simmel believes his pragmatic-naturalist theory of truth entitles him to a pragmatic-naturalist standard of truth as well. The two, however, do not necessarily entail one another. The criterion of truth only establishes the test necessary or simply possible for determining the truth of a proposition, not what the nature of truth itself is. Simmel insists that “we do not have any other defi nitive criterion for the truth of a represen-tation except that the actions instituted by it yield desired consequences” (108, translation altered; 103). This claim, however, appears to fl y in the face of his earlier response to the Dilemma of the Criterion in which he took the standard of truth to be inferentially related items of knowledge, that is, epistemic coherence.

The inconsistency that emerges from the two competing criteria of truth is potentially resolved when Simmel argues that “once these modes of representation have been fi nally established as expedient through selection and cultivation, they form among themselves a realm of theory that deter-mines, according to inner criteria, the inclusion or exclusion of every new representation” (108; 103). Illustrating this statement, he appeals to the “axioms and methodological norms” from which all geometrical principles are derived but which are themselves valid only “in relation to something external, such as the nature of space, our mode of intuition and the compul-sion of our cognitive norms [ Denknormen ]” (108, translation altered; 103).

I submit that in this passage Simmel unites the pragmatic-naturalist and the inferentialist standards of truth. The fundamental “norms and facts” of human knowledge are established in virtue of their function in guiding actions that help to maintain and promote human life. This is a process of trial and error in which beliefs are tested against the struggle



to sustain life and develop its potentials. At some point, these basic beliefs congeal into a separate “realm of theory”—much like Marx’s description of the historical separation between physical and intellectual labor—which then employs its own “inner criteria” to establish the veracity of new beliefs. I contend that the “inner criteria” to which Simmel refers is nothing less than the inferentialist doctrine that holds that the truth of a proposition can only be known according to an inferentially related proposition taken as a “criterion of truth.” This means that propositions or beliefs enter into the inferential system of knowledge, whether it be conceived ideally as a circle or an infi nite regress, as a consequence of leading the organism to success-ful actions, where “successful” means that actions produce their intended consequences. Once this pragmatic criterion has secured a certain amount of propositions or “modes of representation,” these items of knowledge are able to function normatively to incorporate new items according to what can be inferred to and from them. When a new claim is found to cohere inferentially with the already established set of knowledge, then this claim is said to have been “demonstrated” and subsequently becomes a new item of knowledge.

One essential point to consider in this description is that in order to work, the connection between propositions in the “realm of theory” must be one of “demonstration,” which does not include pragmatic proof. For Sim-mel, demonstration, no matter how broadly or narrowly construed, always refers to inferential relations between propositions or beliefs. Since prag-matic proof relates beliefs to action and action to natural drives, it can never constitute demonstration in the inferentialist sense. In other words, Sim-mel has offered two divergent modes of verifi cation: one pragmatic-natural-ist and the other inferentialist (this being common to both of the responses to the Dilemma of the Criterion). So what, we might ask, is the relationship between the two?

Prima facie, the relation Simmel intimates is a primarily genetic one, that is, a description of the historical shift in verifi cation from pragmatics to inferential demonstration. Since Simmel never explicitly identifi es the subject of this history, the process may refer to either the individual (onto-genesis) or the human species in general (phylogenesis). In other words, the shift from quantitative accumulation of pragmatically instituted beliefs to the inferentialist mode of verifi cation might refl ect either a transition in human history or a development that occurs in the psycho-epistemic life of every individual.


jared a. millson196

Still, Simmel’s comparison of the relation between the realm of prag-matic verifi cation and the theoretical domain of inferentialist verifi cation to the role of particular geometrical judgments within an entire geometri-cal system illuminates a normative or epistemological account paralleling the genetic one. The difference between the two kinds of verifi cation depends upon what one takes as the subject matter of verifi cation. If that subject matter is an individual judgment, then the relevant criteria are other items of knowledge, and hence the epistemic norms are inferences. But if the subject matter in question is the totality of these presuppositions, then the pragmatic-naturalist model offers the only possible mode of criteria. Sim-mel thus avers that “individual cognitions may mutually support one other, since the norms and facts already established validate others, but the totality of these norms and facts has validity only in relation to determinate physio-psychological organizations, their conditions of life and the furthering of their activity” (108, translation altered; 103). Questions persist, however. When, for instance, are we even in a position to ask about the validity of all our true beliefs? If pragmatic verifi cation is even possible, is it not the case that our knowledge of human physiology and psychology is itself con-stituted by claims in need of inferentialist verifi cation? Finally, what is the relation between the genetic and normative accounts? Does one take prece-dence over the other?

These questions draw attention to characteristics that make the prag-matic-naturalist conception of truth and the genetic account of developing verifi cation processes distinctly relativistic . Under the pragmatic-naturalist model, truth is relative in the sense that it is determined by the physio-psychological constitution of the human species and the conditions neces-sary for the maintenance of human life. As Simmel notes, other species with different physical and psychological organizations will have different representations, different needs, and different practical engagements with the world. This position is conducive with our ordinary view of relativism since the truth of any proposition is relative to a particular “perspective,” in this case a natural species. Understood in these terms, Simmel has pre-sented a type of relativism that, though elaborately articulated, ultimately fi ts with the traditional concept of relativism. His “genetic” account of the relationship between pragmatic and inferentialist verifi cation is also con-ducive with this view. According to that interpretation, the proper mode of verifi cation and hence the very nature of truth would be relative to our place in the development (either ontogenetic or phylogenetic) from one to



the other. As we shall see, however, what appears to be an affi rmation of traditional relativism is in fact its radical critique, and the questions raised by the second and third formulations are answered in his fourth.

Heuristics

Simmel’s fourth formulation of relativism as heuristic relativism repre-sents a point of view that I believe not only makes possible the reconcili-ation between pragmatic and inferentialist models of verifi cation but also makes explicit his epistemic theory of truth. Most of all, it achieves this by employing the very element of refl exivity that grounds the charge of relativ-ism’s self-refutation.

In this formulation, Simmel asserts that relativism means that “the con-stitutive principles that claim to express, once and for all, the essence of objects are transposed into regulative principles which are only points of view in the progress of knowledge” (110; 106). He goes on to elaborate the advantage of this approach:

If the constitutive assertions that aim to establish the essence of things are changed into heuristic assertions that seek only to determine our methods of attaining knowledge by formulating ideal ends, this makes possible the simultaneous validity of opposing principles. If their signifi cance is only methodological, they may be used alterna-tively without contradiction; there is no contradiction in changing from the inductive to the deductive method. The true unity of appre-hension is secured only by such a dissolution of dogmatic rigidity into the living and moving process. Its ultimate principles become realized not in the form of mutual exclusion, but in the form of mutual dependence, mutual evocation and mutual complementation in practice. (110, translation altered; 107)

Simmel believes that he has found in this proposed translation of constitu-tive judgments into regulative or heuristic principles the key to a sustain-able relativism. Foundational a priori principles do not constitute the objects of our empirical knowledge. Rather, they form methodological guides for our verifi cation processes. They are normative in the sense that they tell us what we ought to do given certain conditions and purposes. Their “truth”


jared a. millson198

lies in their ability to perform this function. This claim thus resonates with Simmel’s fi rst formulation, which treated foundational knowledge claims as provisional bases for verifi cation.

However, this new heuristic formulation adds the claim that determi-nant principles derive their own validity from a reciprocal relation with their opposite. Following one heuristic principle leads one to presuppose the validity of its opposite in a movement that resembles the hermeneutic circle. To cite but one of Simmel’s examples:

Nothing prevents us from attempting to trace any merely given state of the world to the mental conditions that have produced it as a con-tent of representations; just as nothing stands in the way of tracing these conditions to cosmic, historical or social facts which could give rise to a mind equipped with these powers and forms. The image of these facts, external to the mind, may again be derived from the subjective presuppositions of natural-scientifi c and historical knowl-edge, and these again from the objective conditions of their genesis, and so on ad infi nitum . Of course, this knowledge is never realized in a pure schema; rather the two tendencies commingle in a fragmen-tary, discontinuous and contingent way. But the transformation of both into heuristic principles dissolves their opposition into reciproc-ity and their mutual negation into an endless process of reciprocal activation. (113, translation altered; 111)

Taken only as a heuristic principle, modern idealism’s claim to derive the world from the structure of human cognition is no longer opposed to a materialism that fi nds the origin of this structure in historical, social, or physical facts. On one hand, Simmel is achieving with his heuristics what most relativisms claim to provide in their reduction of knowledge and truth to perspectives: a resolution of the contradiction that arises when two opposed propositions are held to be true at once. On the other hand, Simmel is not merely salvaging simultaneous validity; he is also arguing that qua heuristic principles, opposing propositions depend on one another. The nature of this dependence is not, however, formal-logical but, rather, methodological, material, and semantic.

I believe that the type of interaction and mutual dependence Simmel has in mind here is much like that of the hermeneutic circle: understand-ing any part of a whole requires contextualization that can only come about



on the basis of an understanding of the whole, and yet this understanding is itself dependent on a recognition of the constituent parts and their rela-tions that make up the whole. The hermeneutic circle is a way of explaining and expressing the understanding and interpretation of a work of art as a continuous process that unfolds in time. Interpretation gradually changes to incorporate new information that is acquired in one perspective by switching to its complementary mode. The process may be viewed both as progressive and as infi nite, but the relation between approaches is one of methodological and semantic dependence and support, not strict logical entailment. That is, comprehending the constituent parts of something is only possible if one also comprehends the whole and vice versa. The alter-nating principles provide the conditions for the possibility of understanding each other. Following Robert Brandom, we can call this kind of relationship a “reciprocal sense -dependence.” 6 Understanding what one claim means demands that we understand what the other means. This is a semantic or “sense” relation, not an ontological one, because it does not entail the existence of objects to which the statements refer. In other words, the rela-tionship is not one of “reciprocal reference dependence.” It is precisely the latter type of relation that Simmel wants to abandon when he talks about relinquishing the idea of constitutive principles.

In this sense, we may interpret, following one of Simmel’s examples, the relation between idealism and materialism as one of “sense reciprocity”: understanding the world of facts as the product of human cognition can only proceed by in turn understanding the nature of human cognition on the basis of facts, and these facts again in terms of their origin in human cognition. Now, one objection that might be leveled against this claim is that such reciprocity confl ates two different kinds of causality: effi cient and transcendental. According to Kant, the synthetic activity of the mind only constitutes empirical objects as objects of empirical cognition; it is not the real, effi cient cause of objects in the world, though effi cient causality is among the transcendental categories by which the mind necessarily under-stands these objects. Materialism, on the other hand, conceives of the mind as the effect of certain material conditions that serve as its effi cient cause. I believe that Simmel would reply to this objection by suggesting that a strict separation between effi cient and logical/transcendental causality, or between genesis and validity more broadly, presupposes that each describes a determinant objective relationship. If the propositions are conceived as heuristic principles, as methodological guides, then their hypostatization is


jared a. millson200

reversed and the possibility is opened for an interaction aimed at mutual comprehension. Such comprehension would involve a conception of the world as intelligible in virtue of a priori structures of human cognition that in turn depend upon the very empirical objects and individuals they make possible. Actually, I have already indicated some of the details of this posi-tion when I highlighted Simmel’s claim that a priori structures in general may be conceived as determinant of the empirical world while admitting that any particular a priori must be amenable to empirico-genetic reduction (see above).

Another objector may charge that even if Simmel’s heuristic approach to truth claims is capable of reconciling opposing propositions, it can never establish the objectivity of these propositions because they are only subjec-tive principles that bear no normative force. Simmel would indeed agree that qua heuristic assertions, opposing propositions are only guidelines for subjects to follow in their search for knowledge and are therefore relative to the nature and existence of human cognitive subjects. However, he would reply that “though each of these methods remains to some extent subjec-tive, yet together, through the relativity of their application, they seem to express adequately the objective signifi cance of things” (114; 113). Just as for Kant objects of empirical cognition are constituted by the relations of subjective impressions, so too do subjective heuristic principles achieve an objective validity through their interaction and reciprocity. This “relativity” is not the “additional qualifi cation of an otherwise independent notion of truth but is the essence of truth itself” (116, translation altered; 116). In other words, it is precisely in virtue of their reciprocal relations with their contraries that truth claims attain their objective, normative bindingness.

Someone else might argue that even heuristic principles make irreduc-ible claims to objective truth. Regulative assertions may claim legitimacy by appealing to the necessary structure of human cognition or to the nature of knowable objects. They may insist that given their particular congru-ence with subjective or objective features of cognition, they are uniquely suited to the task of knowledge acquisition. Indeed, at one point even Sim-mel appears to conceive of regulative principles in a Kantian fashion as incognizable ideas demanded by the structure of human understanding and the excessive tendency of reason. 7 However, in order to remain con-sistent, Simmel should reply that the appeals of regulative principles to “higher” constitutive ones merely require a further application of heuristics to these upper-order truth claims. The validity of a heuristic principle does



not lie in its correspondence with any fundamental reality, either subjective or objective, but only in its interaction and “relativity” shared with another heuristic principle.

This hypothetical Simmelian response brings forward one fi nal and crucial objection. As I noted in the beginning of this article, the classic objec-tion to relativism is proposed through a refl exive application of the rela-tivist principle that supposedly reveals its self-refutation. A similar charge may be leveled at Simmel’s heuristic relativism. Insofar as it demands that all constitutive truth claims are to be translated into heuristic principles that derive their validity from a reciprocal relation to and interaction with other principles, does not this assertion itself have the status of a heuristic principle? Simmel’s answer is affi rmative: “Heuristics, which is only the consequence of the application of the relativistic principle to the categories of knowledge, can accept without contradiction that it is itself a heuristic principle” (117; 117–18). Likewise, relativism’s validity is relative to its inter-action and alternation with its opposite. Simmel writes:

Only a relativistic epistemology does not claim exception from its own principle; it is not destroyed by the fact that its validity is only relative. For even if it is valid—historically, factually, psychologically—only in alternation and harmony with other absolute or substantial principles, its relation to its own opposite is itself only relative. . . . The question as to the grounding of the relativistic principle which is not incorporated in the principle itself, is not ruinous for relativism, because the ground is removed to infi nity, i.e. it strives to dissolve every absolute that presents itself into relation, and proceeds in the same way with the absolute that offers itself as the ground for this new relation. This is a process which knows no stopping point and whose heuristic sublates the alternative: either to deny or to accept the absolute. It makes no difference how one expresses it: either that there is an absolute but it can be grasped only in an infi nite process, or that there are only relations but they can only replace the absolute in an infi nite process. (117, translation altered; 118)

In order to apply its own claims to itself, relativism must be true only in relation to absolutism; yet this relation itself is not to be hypostasized but recognized in turn as a heuristic assertion that has validity only rela-tive to its own opposite. The last part of the passage reiterates Simmel’s


jared a. millson202

conclusion that the two outcomes from the Dilemma of the Criterion (foun-dationalism and coherentism) are mutually compatible as ideal possibili-ties. If relativism is construed as the coherentist or circularist claim that truth is a function of relations among items of knowledge and absolutism is read as the foundationalist claim that truth is a characteristic of self-evi-dent propositions whose verifi cation nevertheless depends upon inferential “demonstration,” then relativism and absolutism are mutually compatible. If “knowing the truth” is always a matter of inferential verifi cation, then it makes no difference whether truth consists in those very inferential rela-tions or if it persists wholly outside and untouched by them. The possibility of knowledge having its basis in an absolute truth haunts the coherent-ist just as the possibility of circular reasoning (i.e., of returning to one’s original claim) haunts the foundationalist. In a wholly inferentialist epis-temic process, the two conjectures function as guiding ideals, as heuristic principles, which project possible structures of knowledge. In other words, heuristics treats every knowledge claim, including metaphysical theories of truth , as methodological norms for processes of verifi cation. This means that the metaphysical theory of absolute or necessary truths, when taken heuristically, is nothing more than Simmel’s fallibilistic foundationalism. It is an injunction regarding how one ought to think of truth and how one ought to proceed in inquiry. Our ability to deploy such a heuristic principle depends upon our ability to understand its opposed, coherentist theory of truth, insofar as we search for absolute starting points as a means to avoid circularity. Since any theory of truth is only as good as its methodological function in our concrete activities of inquiry, and since both foundational-ist and coherentist theories prescribe the same course of action in these inquiries, that is, treating certain cognitions as a priori until their own con-ditions can be determined, the two theories are equally compelling. In any case, the payoff of heuristics is a theory of truth that is distinctly epistemic. Heuristics is able to survive self-refl exive application because it itself func-tions as a stipulated guide to inquiry.

Moreover, once interpreted heuristically, epistemology itself becomes a hermeneutic discipline, characterized by the reciprocal dependence between theories of truth and theories of knowledge. The very distinction between the nature of truth and the standard of truth breaks down when both are conceived heuristically. Simmel is right: according to his brand of relativism, the choice between affi rming or denying the absolute is obviated; absolut-ism is merely placed alongside relativism in an infi nite reciprocity of truth



claims. Any attempt to escape or halt this process of interaction is forced to present itself as a truth claim and thus submit itself to the very infi nity it wishes to evade. In this sense, Simmel is quite perceptive when he claims that “relativism is closer than one would think to its extreme opposite— Spinoza’s philosophy” (118; 120). A monism in which the absolute includes all that exists dissolves itself into an infi nite system of relations and processes from which nothing can escape. Ultimately, the heart of Simmel’s refl exive relativism lies in the self-applicable declaration that no truth claim can be absolutely self-suffi cient in either a psychological or a logical sense, and, of course, psychology and logic are themselves opposed heuristic principles whose validity derives from mutual interaction and alternation; it is, in fact, in this alternation and infi nite reciprocity that truth lies.

With this fi nal formulation of relativism and its response to the prob-lem of refl exivity, we are fi nally in a position to reassemble the remaining elements of Simmel’s epistemology. By rendering the pragmatic-naturalist theory of truth into a heuristic principle, this theory serves simultaneously as a model of verifi cation and as an ideal conception of the truth toward which verifi cation is aimed. However, two problems remain in the attempt to reconcile the pragmatic-naturalist account with the later formulation of heuristic relativism. The fi rst of these lies in the fact that pragmatic- naturalism looks very much like a standard relativism of perspectives sus-ceptible to the self-refutation argument—in this case, truth and the criterion of truth are relative to the physiopsychological makeup of the human spe-cies. Appearances aside, I believe that there is a way for Simmel to avoid this conclusion. Just as idealism and materialism were shown to be mutu-ally supportive heuristic principles, so too can representations and the facts of human anatomy. While verifying the truth of any representation may involve tracing the actions it inspires back to the conditions for human life, verifying the truth of these biological facts depends upon our use of representations. In this sense, the criterion of truth is relative to the phys-iopsychological constitution of the human species, which is relative to the representations operative in human cognition, which are again relative to human biology. This hermeneutic spiral does not paralyze verifi cation but, instead, makes possible the progress of knowledge by unfreezing thought and preventing hypostasis.

The second problem that arises is that of connecting pragmatic-natural-ist and coherence standards of truth. As we saw above, the relation between the two is presented historically or genetically as well as normatively or


jared a. millson204

epistemologically. Historically, we began with the pragmatic-naturalist stan-dard of truth, and once enough verifi ed beliefs were accumulated by this procedure, we developed an inferentialist criterion of truth. (Remember that the problem of moving between a theory of truth and a standard of truth has been obviated by Simmel’s heuristics.) Again, the subject referred to here may be either an individual or the human species itself. Either way the problem seems to be that our standard of truth is relative to which stage in the genetic story we happen to fi nd ourselves in.

The secret to avoiding the typical dilemma of relativism lies, according to Simmel, in taking up a self-refl exive position and driving it to its radi-cal extreme. Stepping back from the genetic account, we can see that any particular proposition in this argument—say, the biological claim about the unique physiological characteristics of human beings—is verifi able only by articulating its inferential relations to other items of knowledge, that is, only according to the inferentialist criterion. In other words, coming to know the truth of the claims deployed in the genetic account, including the determination of which “stage” we are in, will invariably be a function of inferentialist verifi cation. Conversely, understanding what it means to verify something according to our background knowledge will require us to offer a genetic explanation of our knowledge acquisition. This means that genesis and validity, historical and normative accounts, are reciprocally sense dependent. Any particular truth claim will be verifi ed inferentially, but in order to explain how it is that we have knowledge in general, we must offer a genetic account that proceeds from pragmatic-naturalist criteria to inferentialist ones.

This reciprocity between genesis and validity of knowledge intimates Simmel’s refl exive response to the challenge of relativism in general. For Simmel, the restriction of truth or knowledge to a specifi c existential “per-spective” (society, epoch, culture, individual, species, etc.) is not a means to silence endless dispute or to recognize particularity and difference. Instead, it is an impetus to refl ection. The reduction of knowledge and truth to a “perspective” is not the end but the beginning of inquiry simply because this calls for an investigation of one perspective that inevitability leads to that of others; indeed, our knowledge of one class of perspectives is itself perspectival. One way of thinking about this idea involves recognizing that there are empirical facts that provide the evidence for the relativist’s fi rst-order , descriptive claims (e.g., “There are and have been many different human societies, and each has its own notion of what counts as true, etc.”),



which in turn form the basis from which second-order , normative assertions (e.g., “Truth is relative to each society”) arise. Simmel’s suggestion is that fi rst-order claims are “reciprocally sense dependent” upon the second-order ones. Verifying descriptive claims is dependent upon one’s understand-ing of what counts as normative in verifi cation. Conversely, methodologi-cal norms accrue naturalistically as part of a historical process. In other words, understanding the content of norms operative in inquiry depends upon one’s ability to grasp the historical, genetic, or pragmatic-naturalistic process by which such norms acquire their binding force.

What Simmel’s heuristic relativism demonstrates is that if relativism is rigorously submitted to its own conclusions, then the very “topic-per-spective” structure that defi nes relativism as such begins to break down. “Topics” are relativized to “perspectives,” which are in turn relativized to “topics”; normative categories are reciprocally sense dependent upon empirical facts. However, lest this reciprocity be interpreted as evidence of Simmel’s capitulation to a hyperbolic brand of idealism, it must be made clear that the distinction between “topic” and “perspective” is never effaced. Perspectives are never referentially dependent on empirical claims that might be inserted back into the circular reciprocity among heuristic principles; they always refer to a realm of nature and action that is existen-tially independent of the semantic realm of propositions, judgments, and representations.

The solutions found for each of the problems that arose in our dissec-tion of Simmel’s previous three formulations demonstrate not only that his relativism is refl exive among its multiple aspects and levels but also that it is not synonymous with theoretical egalitarianism. Simmel may believe that all elements of human knowledge depend on each other for verifi ca-tion; this does not commit him to the belief that all are equally valid. There are certain propositions, positions, and phenomena that give privileged access to the reciprocity of elements. Simmel’s formulations of relativism are four such points of entry into the interconnected world of experience and knowledge. They each sustain a duality that gives them a special sig-nifi cance in epistemological inquiry: on one hand, they reveal the recipro-cal relations among truth claims; on the other hand, they are themselves bound up within these relations as just one claim among others. Money too shares this dual structure, which “consists, on the one hand, in measuring the value relations of goods exchanged and, on the other hand, in being exchanged with these goods and thus itself becoming a quantity subject


jared a. millson206

to measurement” (122; 126). Otherwise put, both relativism and money can be included among “those normative ideas that obey the norms that they themselves represent” (122; 126). The relationship of relativism to the methodology and substance of social inquiry that becomes apparent in this defi nition of money ought to remind us that my attempt to distill the theoretical signifi cance of Georg Simmel’s epistemological relativism is necessarily incomplete so long as it remains abstracted from his broader program of concrete sociological research. With this analysis and recon-struction, I hope to have provoked an interest in pursuing the larger task of integration among the epistemological, methodological, and sociological components of Simmel’s philosophy.

notes

1. Socrates’ argument in the dialogue takes up a number of different topics but generally shares the same self-referential structure, that is, it continually applies the claims of relativists or “philosophers of fl ux” to themselves. The best instance of the argument for my more specifi cally epistemological concerns is found in Theaetetus 170b–171c.

2. Georg Simmel, The Philosophy of Money , ed. David Frisby, trans. Tom Bottomore and David Frisby (New York: Routledge, 2004); Georg Simmel, Philosophie des Geldes. Gesamtausgabe Band 6 (Frankfurt am Main: Suhrkamp, 1989). Quotations are cited parenthetically by page number in the text, listing fi rst the translation and then the original.

3. The problem of criterion is central to Sextus’s account of Pyrrhonian skepti-cism and thus appears numerous times and in various forms throughout his works. Important expositions of the problem appear in Against the Logicians , trans. Richard Bett (Cambridge: Cambridge University Press, 2005), bk. 1, 314–19; and in Outlines of Skepticism , trans. Julia Annas and Jonathan Barnes (Cambridge: Cambridge University Press, 2000), bk. 1, 114–17. For a treatment of the problem of criterion in a contemporary epistemological light, see Ruth Weintraub, The Skeptical Challenge (London: Routledge, 1997), 77–87.

4. A good source for an overview of these theories is Nicholas Rescher’s classic The Coherence Theory of Truth (Oxford: Clarendon Press, 1973).

5. William James, “What Pragmatism Means,” in William James: Writings 1902–1910 (New York: Library of America, 1987), 520. James’s pragmatic concep-tion of truth shows other important features of Simmel’s relativism. Both, for instance, conceive of truth as inherently inferential, functional, and processual. See David C. Lamberth, William James and the Metaphysics of Experience (Cambridge: Cambridge University Press, 1999), 48–56.



6. Robert Brandom, Tales of the Mighty Dead: Historical Essays in the Metaphysics of Intentionality (Cambridge: Harvard University Press, 2002), 194–97. Brandom defi nes sense dependence as follows: “Concept P is sense dependent on Concept Q just in case one cannot count as having grasped P unless one counts as grasping Q” ( Tales of the Mighty Dead , 194).

7. He claims that a reliance on constitutive assertions is to be replaced by “the notion that our understanding must proceed as if things behave in such and such a way” (110; 106).


The Reflexive Relativism of Georg Simmel

Documents

Transcript of The Reflexive Relativism of Georg Simmel