Tales of the mighty tautologists? Frank Scalambrino

Author Information: Frank Scalambrino, University of Dallas, Texas, scalambrinof9@gmail.com

Normative Functionalism and the Pittsburgh School: Table of Contents

Scalambrino, Frank. 2012. “Tales of the mighty tautologists?” Social Epistemology Review and Reply Collective 2 (1): 83-97.

The PDF of the article gives specific page numbers. Shortlink: http://wp.me/p1Bfg0-Ca


There is supposed to be deep agreement among the work of Wilfrid Sellars, Robert Brandom, and John McDowell in regard to normativity. As a result, according to Robert Brandom (2008), and echoed by Chauncey Maher (2012), “normative functionalism” (NF) may refer to a position held by Sellars, Brandom, and McDowell, i.e., “The Pittsburgh School” of philosophy. The standard criticism of the various forms of this normative functionalist position points out the inconsistency in the commitment of normative functionalists to both metaphysical realism and psychological nominalism. Yet, the inconsistency between metaphysical realism and psychological nominalism may be difficult to see until the relation between normativity and perception is clarified. To this end, in this article I discuss the role of habit in perception. Normative functionalists aspire for a sort of pragmatism between the horns of psychologism and pan-logicism. However, once a discussion of habit in perception reveals a kind of relation between an agent and its environment that exceeds the inferential capacity of normativity, the normative functionalist position seems tautological. Put more specifically, the NF thesis may merely be claiming that the inferential sort of normativity which governs rational synthetic processing of experience is an inferential sort of normativity governing rational synthetic processing. The revelation of such a tautological grounding should be sufficient evidence for the Pittsburgh School to consider re-working its understanding of the functionality of normativity; for example, regarding claims such as: “In an important sense there is no such boundary [between the discursive and non-discursive], and so nothing outside the realm of the conceptual” (Brandom 2000, 357). This discussion should be, at least, valuable as a supplement to the standard criticism of NF or in regard to the Pittsburgh School’s avowed relation to G.W.F. Hegel.

“[A] tree or a rock can become subject to norms insofar as we consider it as engaging in social practices.” [1] — Robert Brandom

I. Introduction

The “Pittsburgh School” of philosophy refers to the work of Wilfrid Sellars, Robert Brandom, and John McDowell from the University of Pittsburgh. And, there is supposed to be “deep agreement” within the Pittsburgh School regarding normativity (cf. Brandom 2008, 357; cf. Maher 2012). “Normative Functionalism,” then, refers to the philosophical position indicated by the deep agreement among these various Pittsburgh School understandings of normativity. So, how may the position of normative functionalism (NF) be characterized?

Consider Brandom’s characterization from his Reason in Philosophy: Animating Ideas (2009),

The synthesis of a rational self or subject: what is responsible for the [normative] commitments … has a rational unity in that the commitments it comprises are treated as reasons for and against other commitments, as normatively obliging one to acknowledge some further commitments and prohibiting acknowledgement of others [Brandom’s emphases] (Brandom, 2009, 14).

It is as if the meaning of an experience for an agent depends on its relation to the norm-governed network, i.e., a space of reasons, in which it functions as an assertion. And, according to Brandom: “This is Kant’s normative inferential conception of awareness or experience” (Brandom 2009, 14). Further, in his book The Pittsburgh School of Philosophy (2012), Chauncey Maher explains, “the big idea is that the meaning of a term or a whole sentence is its norm-governed [emphasis added] role in rational conduct, broadly construed to include perception, thinking, speech, and deliberate action” (Maher 2012a, 5). Ultimately, in this article, I will argue that the domain of experience the Pittsburgh School considers norm-governed is too widely construed in regard to perception.

On the one hand, this criticism is not novel. Jürgen Habermas,[2] Joseph Margolis, Tom Rockmore, James Swindal, Patrick Reider, and Wilhelm Wurzer, among others, have in various ways already criticized NF for what may be called its commitment to “metaphysical realism.” So, for the purposes of this article, call this critical position that of “The Other Pittsburgh School”, i.e., Duquesne University. On the other hand, the focus of my discussion regarding NF centers on the role of habit in perceptual experience. And, this focus is an innovation.

The kind of perceptual memory I am concerned to introduce into the conversation corrects the problem regarding the overly wide construal of NF. Rather than hold a position resembling the claim that perception is conceptual “all the way out to the impressions” (McDowell 1994, 69), the kind of memory I will discuss is non-discursive and involves relations. So this kind of memory is non-conceptual because it is relational, not because it has non-conceptual content. In other words, to question whether non-discursive habit has conceptual or non-conceptual content indicates a longstanding prejudice among philosophers regarding memory and fails to appreciate this kind of perceptual memory. Hence, in this article I advocate for incorporating the differentiating function of non-discursive memory into the research project of NF. Whereas criticism from the Other Pittsburgh School tends to invoke G.W.F. Hegel’s Phänomenologie des Geistes (Phenomenology of Spirit) and illustrate the incommensurability between two of the normative functionalist’s commitments, i.e., metaphysical realism and psychological nominalism, I invoke Immanuel Kant’s Kritik der reinen Vernunft (Critique of Pure Reason) and illustrate how the Pittsburgh School’s understanding of perception as norm-governed misunderstands the role of habit in perceptual experience. That is to say, a non-discursive memory, i.e., habit, governed synopsis buffers the relation between the rational synthesis of norm-governed aspects of experience and the mind external environment. Due to the limited space of this article, I will focus primarily on Chauncey Maher’s claims in his book The Pittsburgh School of Philosophy and the work of Robert Brandom.

Specifically, I argue that the Pittsburgh School’s thesis of normative functionalism does not consider the function of procedural memory involving perceptual relations, i.e., non-discursive habit, to differentiate which aspects of the environment are instantiated into lived perceptual experience. And, I then show how including this differentiating function of non-discursive habit necessitates that the Pittsburgh School reduce the domain of experience to which NF is supposed to pertain. In other words, the Pittsburgh School must acknowledge habit-governed aspects of experience in addition to norm-governed aspects, and as non-conceptual, these habit-governed aspects are beyond the purview of normativity. Lastly, I show that though acknowledging habit-governed aspects of experience would allow normative functionalists to adjudicate themselves from the Other Pittsburgh School’s accusation of a commitment to metaphysical realism, it would also follow that the normative functionalist position is ultimately tautologically grounded.

That the tautological grounding of NF is undesirable may be seen in two primary ways. First, it leads normative functionalists to claims such as Brandom’s: “In an important sense there is no such boundary [between the discursive and non-discursive], and so nothing outside the realm of the conceptual” (Brandom, 2000, p. 357). Not only is such a claim false, it hinges on a misunderstanding of the concept of “nothing” (cf. Scalambrino, 2011). Second, regarding the mind external environment the totality of the norm-governed space traversable by inference would necessarily be limited by communal reciprocity, i.e., agreement, among adherents to the norms. Put another way, through projective anticipation a map may be used to guide the re-structuring of a terrain. Yet, originarily maps depend on terrains, not the other way around; and, as a result terrains always exceed the capacity of mapping techniques to represent them — mapping techniques being one manner of relating to terrains.

Hence, my discussion of perceptual memory in this article should motivate the Pittsburgh School to re-work its understanding of the difference between discursive practices and, what Brandom calls “the non-concept-using world” in which these practices occur (cf. Brandom 2000, 356). There is, of course, a difference between claims regarding what can be done with (that which is) the mind external environment and claims regarding what the mind external environment is. And, in Kant’s wake metaphysical realism is a bust. In other words, the work of the Pittsburgh School does not overcome the paradigm shift invoked by Kant’s Copernican revolution. Perhaps this insight regarding metaphysical realism may be also gained through the later Ludwig Wittgenstein’s notion that language games depend upon forms of life (cf. Wittgenstein 1969, 62 §475) and Thomas Kuhn’s notion of incommensurability between paradigms (cf. Kuhn 1962, 148-150), among others.

II. The collapse of the discovery / demonstration distinction

The discovery/demonstration distinction is at least as old as Aristotle, and it is retained in Kant’s vocabulary through his distinction between regulative and constitutive ideas (A 508/B 536-A 567/B 595). The transparency resulting from ideas, i.e., thoughts, playing a role in both the dimension of discovery and the dimension of demonstration does not mean that the logic of the relations of the one is the logic of the relations of the other. For example, both dimensions, i.e., thought and knowledge, include assumptions; yet, as their points of departure reveal, the demonstration dimension relates to an inferential cluster of norms for the purpose of justification, and the dialectic dimension of thought relates in discovery to information. Elsewhere I discuss more thoroughly the relation between the discovery/demonstration distinction and Aristotle (Scalambrino 2011); suffice to say here that Aristotle explains this distinction in the Prior Analytics (23b24-24a34).

In both his book regarding NF, i.e., The Pittsburgh School of Philosophy, and his “Reply to Reider” Maher uses the same example to illustrate a fundamental point regarding NF. Maher tells the story of “Sarah and Mark.” According to Maher,

Mark and Sarah enter their house, walk into the kitchen to put away some groceries, and Sarah recalls that the air conditioning is on. She asks Mark whether the door is shut. He looks and sees that it is shut and says so, thereby passing his knowledge on to Sarah, whom we would normally now count as knowing that the door is shut (Maher 2012a, 7-8).

In regard to this story Maher suggests that the search to justify Sarah’s knowledge leads to a problem. He then goes on to elaborate the worry of the “regress problem.” Namely that “The problem here should be clear: this demand for justification seems like it can go on forever” (Maher 2012a, 5). Further,

This worry is not unique to Sarah’s case; it is a general worry. The fact that knowledge requires justification seems to spark a regress of reasons or justifications. If it doesn’t stop somewhere, there would seem to be no knowledge at all. Thus it can seem like we need a foundation of knowledge, something that will halt the regress, something that one can know without any further reasons or knowledge (Maher 2012a, 7-8).

I quote Maher at length here to show that the enunciation of NF for the Pittsburgh School results as a solution to the regress problem. Yet, as I indicated with the title of this section, the Pittsburgh School’s articulation of the regress problem does not seem to appreciate the distinction between discovery and demonstration. Maher’s use of the word “knowledge” makes this clear.

Specifically, when Mark, in the above example, sees what appears to be a closed door, he does not have knowledge that the door is closed. Rather, he has discovered evidence that might be used to demonstrate that the door is closed. Therefore, he cannot pass on “knowledge” to Sarah that the door is closed. So, in hearing Mark submitting his evidential testimony toward demonstrating to Sarah that the door is closed, Sarah discovers evidence with which she might persuade herself to believe the door is closed; that is to say, she now has evidential testimony with which to decide whether she believes the closure of the door to have been sufficiently demonstrated for her.

Further, regarding Mark, Maher claims, “although Mark’s knowledge is non-inferential, it is not independent of other knowledge” (Maher 2012a, 9). Maher goes on to unpack the dependence of Mark’s knowledge on “other knowledge” to emphasize that since language users are following rules, experience must conform to these rules, i.e., be norm-governed. In fact, according to the Pittsburgh School, “One cannot talk or think unless one sufficiently complies with ‘the rules’” (Maher 2012a, 4). Yet, such Pittsburgh School realizations apply to the dimension of demonstration. If Maher were to not use the term “knowledge” regarding what is merely a hypothesis for Mark, then the Pittsburgh School’s confusion would be cleared. And, it would seem to follow that the scope of NF should be reduced regarding how much of experience is norm-governed upon realizing that the dimension of discovery, i.e., the space of hypothesizing, includes external relations both independent of the rules of language use and non-inferential.

So, in building toward the position of NF, when Maher suggests that something is needed to “halt the regress,” notice he says that what is first needed is something that one can “know.” Yet, were the matter, as the Pittsburgh School would evidently have it, merely about language, then all disputes might be resolved by reference to the norm-governed “space of reasons,” not perceptual experience. What Maher should have said is that agreement regarding the state of affairs in question is what is first needed. Whereas the “normative space” occupied by a group of agents may characterize how they determine a demonstration sufficient, the normative space does not exhaustively determine the dimension of discovery. Because hypothesis, not knowledge, pertains to the dimension of discovery, it is possible for two users of the same language to disagree about the state of affairs in question. For example, gardening need not depend on cartography. In the next section, I discuss the role of habit in perceptual experience by showing how to think about an efficacious habit-governed aspect beyond the function of normativity.

To sum, by collapsing the distinction between discovery and demonstration the Pittsburgh School is able to suggest that norms determine the dimension of discovery just as they determine the dimension of demonstration. However, it is rather the case that sapient beings hypothesize based on what appears in discovery, and sapient beings gain demonstrable knowledge through the normative practice of scrutinizing hypotheses in an inferential space of reasons. Further, though it is possible to claim to have demonstrable knowledge based simply on what appears to an agent through perception, the constraints on what appears through perception are not the same as the constraints in regard to what counts as demonstrated knowledge. In other words, what appears to an agent through perception can result from non-norm-governed relations to the environment. So, what then determines which aspects of the environment are instantiated into perceptual experience beyond the function of normativity?

III. Habit, Pruning, and Agency

There is a difference between the function of habits and the function of norms in the determination of sapient experience. For example, when a skill or manner of relating to the external world is not practiced the capacity to perform the skill or manner of relating can be eliminated. Psychologists and neurologists refer to this process as “pruning.” The principle here being “use it or lose it” (Smilkstein 2003, 59; cf. Craik and Bialystok, 2006). Further, it is widely believed that pruning indicates the diminishment of capacities for the sake of efficiency — much like habit itself. So, since capacities diminish, it follows that the possible events in the external world that an agent can manifest diminish without practice, i.e., without their exercise — again, much like the maintenance of a habit. Hence, the possible relations to the manifestation of events in the external world exceed the agent’s capacity to manifest these events. In other words, numerically there are more possible ways to relate to the external world than what an agent is capable of, even as the agent is entering into its period of highest potency.

Just as different agents can use the same language while cultivating different habits, habits determine the perceptual horizon of, i.e., the instantiation of objects in, experience prior to the function of normativity. This is not to say that norms cannot influence the formation of habits (cf. Maher 2012a, 108). However, if norms were enough, the locution “trying to break a habit” would be meaningless (cf. Habermas 2003, 155). The power of memory in question here is found in the relations, i.e., it is non-discursive. For example, consider that learning occurs even when agents are unable to linguistically express what has been learned (cf. Wolf, Ebeling, and Müller 2009; cf. Erickson 2008). The horizon of possible perceptions is narrowed by the agent’s habitual non-discursive perceptual engagement with that which is external to the mind prior to the judgment of the identity of what is perceived (cf. Sperling 1960). In this way, the differentiating function of habit determines states of affairs by limiting for an agent the ways the agent can perceptually take up an environment.

One should remember, then, to include an account of the agent’s memory when examining the agent’s relation to its external environment. It is as if the Lebenswelt were first and the Umwelt next in the order of awareness. For example, consider the influence of the tendencies of orientation (right or left), preferences for sounds (female or male voices), and preferences regarding color, shape, and symmetry. Norms, even once affirmed, still answer to Scalambrino, the presence of habits, e.g., consider the intention to overcome the habit of smoking. One way, then, to state the alternative to the thesis of the Pittsburgh School’s position of NF for which I am advocating here: habit has a mnemonic grip on the dynamics of perception and that mnemonic grip is governed by actual practice, i.e., the relational activity of practice, not norms (cf. Scalambrino, 2012). An agent must first grapple with habit, not norms, if it is to engage in discursive practices.

Notice that Kant was working toward enunciating these relations with his distinction between constitutive ideas, which pertain to understanding, and regulative ideas, which pertain to reason. In regard to the shift from habit-governed to norm-governed contents, i.e., from dialectic of discovery to justification through demonstration, Kant notes,

we in all cases represented the conditions for their conditioned as belonging to relations of space and time, which is the usual presupposition of common human understanding, on which, therefore, the conflict entirely rested. In this respect all dialectical representations of totality in the series of conditions for a given conditioned were of the same kind throughout … the regress is never thought of as completed, or else, if this were to happen, a member conditioned in itself would have to be falsely assumed to be a first, and hence unconditioned member [emphasis added] … the dynamic series of sensible conditions, on the contrary [emphasis added], allows a further condition different in kind [i.e., habit-governed, not norm-governed], one that is not a part of the series but, as merely intelligible, lies outside the series; in this way reason can be given satisfaction and the unconditioned can be posited prior to appearances [emphasis added] without confounding the series of appearances, which is always conditioned, and without any violation of principles of the understanding (Kant 1998, 530-532 [A 528/B 556-A 532/B 560]).

As this quote shows, Kant consciously avoided the “Myth of the Given” (discussed below) in regard to “the regress” of justifying the identity of perceptual experience. As Kant indicates, what is needed is a “further condition different in kind.” This difference in kind is between conceptual and relational, which is why it does not appear in the series of conditions to which norms apply, i.e., for which we have concepts. With a shift from the series of conditions to “outside the series” and “prior to appearances” Kant moves from constitutive to regulative ideas, i.e., away from normative concepts of the understanding.

As noted in the previous section, the Pittsburgh School seems to want to make regulative ideas constitutive, i.e., to falsely make the normative function of understanding equal to the regulative function of reason’s idea. So, here the notion of a gestalt might provide helpful clarification for the Pittsburgh School: That reason’s idea is greater than the sum of its parts, i.e., what derives from (normatively) taking it apart, means the external world to which it refers is greater than one can understand through an analysis of reason’s idea. Wilhelm Wurzer helpfully refers to this aspect of reason’s relation to experience as “reason’s dehiscence” (Wurzer 1992, 129). In other words, relating an idea of the mind external world to the inferentially constellated space of norms to analyze it will not result in the representation of mind external reality.

It seems better to use the vocabulary of contemporary memory research (cf. Scalambrino 2011), than the language of counterfactual reasoning to discuss the separation, or “dehiscence,” which occurs as reason retracts from considering sensibility to consider the contents of its idea derived from experience. Maintaining the discovery / demonstration distinction should itself show that alternative discoveries to what appears through perception do not depend upon inference; notice, this is the case even if “The limits of my language mean the limits of my world” (Wittgenstein 1922, 149 §5.6). Whereas the demonstration of alternatives may be constrained by the inferentially constellated concepts to which a language provides access, alternative discoveries are constrained by the non-discursive relations that allow for discursive practices. Hence, I suggest this dimension “outside the series” and “prior to appearances” — of the relations between the content of reason’s idea and the other contents which supposedly could have been instantiated through a different consideration of sensibility – is not a constellation of inferentially connected norm-governed ideas; instead, it is a constellation of an agent’s memory-governed capacities to instantiate various perceptual experiences related to the mind external world.

To sum, the contents of perception depend on non-discursive memory in two ways: (1) the relational memory that governs perception, i.e., habit; (2) the identification of the contents of perception requires retentional awareness to fill an experiential role as subject in a judgment of perceptual identification. Now, these mnemonic functions can be influenced through practice and repetition, e.g., forming habits that conform to norms. Yet, even if such habits seem like second nature (cf. Pascal 1995, 33 § 126; cf. Micheyl, et al. 2009; cf. Huys, et al. 2008), neither of these mnemonic functions depends primordially on rationality or normativity, i.e., a “space of reasons.”

So, when Brandom makes the (metaphysical realist) claim that “we are beings who can make explicit how things are” (Brandom 2009, 18), that we can know “how things really are” (Brandom 2009, 17; cf. Brandom 2009, 81, 99, 100, 127; cf. Rockmore 2010, 168), he must be clarified as commenting on how things appear for a group of agents with sufficient agreement (in “form of life”), for to “make explicit” is a kind of demonstration. In other words, a state of affairs may be determined differently by different agents depending upon the manner in which their habits contribute to the determination of their perceptual horizon (cf. Aristotle, 2004, Bk II: 1&2, esp. 1109b); therefore, the differentiating function of this non-discursive perceptual memory points to the dynamics of the dimension of discovery and beyond the aspects of experience, as if mathematically, governed by norms. Relational non-discursive habit determines the perceptual horizon of experience “prior to appearances” and “outside the series” determined by the function of normativity, i.e., the inferentially constellated norm-governed space of reasons.

IV. Tales of the Mighty Tautologists? The Myth of the Given, anti-foundationalism, and the very logic of regress

Is the position of NF ultimately tautological, and if so, then, so what? This section examines the Pittsburgh School’s justification for NF combined with the insights gained from the above discussion. So, despite the Pittsburgh School’s prizing NF as a successful anti-foundationalism that “halts the regress” without relying on a version of the “Myth of the Given,” the regress pertains to the dimension of demonstration, and an anti-foundationalism that avoids the Myth of the Given thereby precludes metaphysical realism. Hence, though a tautologically based grounding of the structure of sapient understanding may have been acceptable to Kant, the Pittsburgh School construal of experience to fall under the purview of tautology would not have been acceptable to Kant.

Put in Kantian terms, the rational synthetic processing of experience moves along a trajectory that functions to limit the dynamic mind external environment through sensibility toward understanding; yet, “prior to” the limiting function of the rational synthetic processing and “outside the series” revealed through rational synthetic processing, perception’s limiting function is always already the experientially-based condition for the possibility of the rational synthetic processing of experience (i.e., post-Locke, Leibniz, and Hume Kant is not Descartes). Ultimately, norm-governed attempts to demonstrate mind external reality must rely on post hoc norm-governed determinations arrived at by traversing the inferentially constellated space of reasons. The Pittsburgh School reading of Kant’s Transcendental Deduction from the Critique of Pure Reason shows where they collapse the distinctions that lead to neglecting the non-norm-governed perceptual limiting of the dynamics of the mind external environment.

The Pittsburgh School builds toward justification of the need for the position of NF to solve the regress problem by discussing what Wilfrid Sellars named “The Myth of the Given” (cf. Sellars1956). Returning to the above example of Mark and Sarah: Maher notes, “it can seem like we need a foundation of knowledge, something that will halt the regress, something that one can know without any further reasons or knowledge. Foundationalism is the view that there is such knowledge” (Maher 2012b, 17). Maher clarifies, “foundational knowledge would have to be knowledge that one could have independently of any other knowledge” (Maher 2012b, 17). And, as such, “The Given is the idea that there could be a foundation of knowledge in that sense” (Maher 2012b, 17). Hence, NF is supposed to provide the (psychological nominalist) anti-foundation that halts the regress without appealing to foundational knowledge, i.e., while avoiding the Myth of the Given.

Yet, it seems important to realize that in the process of objects conforming to mind there can be efficacious non-conceptual aspects of the process on the side of the mind. In other words, avoid accepting the “find a corner in a round room” challenge from the Pittsburgh School that is, at least, traceable to Hegel. For, it is a faulty inference to conclude that: since the concept of non-conceptuality pertains to mind external reality, there is “nothing outside the realm of the conceptual” (Brandom 2000, 357), and therefore mind external reality can be rendered conceptually knowable. Such an inference is what allows for the Pittsburgh School’s inconsistent commitments to psychological nominalism and metaphysical realism, respectively.

As an alternative, I have advocated for incorporating insights from contemporary memory research to help get a better grip on what Kant had in mind. To sum, three concerns should be addressed regarding the Pittsburgh School’s justification of the need for NF. First, recall the Pittsburgh School’s very definition of knowledge requires a connection to other knowledge to count as knowledge. Therefore, the Pittsburgh School has already ruled out the possibility of foundationalism. Second, as I showed above what should “halt the regress” is the weight of the evidence involved in the demonstration of what, if deemed knowledge, would stop the need for further reasons. I offer this as an alternative anti-foundational halt of the regress (more on this below). Third, we must keep separate (a) the normative space of reasons with which we judge the sufficiency of a demonstration to count as justification for deeming a hypothesis “knowledge” and (b) the space of habitual relations to which the perception of a state of affairs is contingent. Keeping these two dimensions separate helps maintain awareness that the state of affairs to which an experiential hypothesis relates is contingent upon an agent’s habit-governed relation to its environment.

According to this alternative then, to say an agent, e.g., Mark, conceives a perceptual norm-governed fact means that the agent conceives of the state of affairs at the time of the event that the fact describes. Notice, hypotheses can function as facts in a space of normative reasons so long as the norms pertain to the state of affairs actualized. Yet, as the above section showed, there are other states of affairs to be considered. Both counter-factual circumspection and counter-factual thinking fail to consider the possibilities relating to the different states of affairs that could have been actualized by different habit-governed environmental relations, e.g., of different (-ly trained) agents.

Moreover, returning to Maher’s example, it seems intuitive to think Mark might respond to Sarah not with evidence of the door’s closure but by trying to open the door or by responding with statements such as: “How good of you to be environmentally conscientious”; “You remind me of something Proust wrote about obsessive types” (cf. Proust 2003, 29-30); or, “Are you trying to get me out of the kitchen so you might hide some of the groceries from me again?” By not accounting for experience in a way that maintains awareness of the contingency of states of affairs to an agent’s habit-governed environmental relation, the Pittsburgh School is ultimately suggesting that states of affairs have a reality independent of (the experiential capacities of) agents. Though the members of the Pittsburgh School differ in how they articulate this metaphysical realist position, Brandom’s way of stating it is: “we are beings who can make explicit how things are” (Brandom 2009, 18). And, the seeds of NF’s inability to keep separate habit-governed and norm-governed spaces can be seen in Brandom’s reading of Kant’s Critique of Pure Reason from his seminal text Making It Explicit: Reasoning, Representing, and Discursive Commitment. [3]

Recall according to Kant, there are three original sources of experience, “namely sense, imagination, and apperception [Kant’s emphases]” (Kant 1998, 225; A 94/B 127). And,

On these are grounded 1) the synopsis of the manifold a priori through sense; 2) synthesis of this manifold through the imagination; finally 3) the unity of this synthesis through original apperception [Kant’s emphases] (Kant, 1998, p. 225 A 94/B 127).

Notice, apprehension as part of the three-fold synthesis is not included with synopsis, but rather appears to coincide with imagination under Kant’s heading “2)”. Further, each one of these grounds has its own law. Whereas you might expect a temporal law regarding sensibility, the law governing synopsis is actually “affinity” (Kant, A 113), and affinity pertains to memory. The law governing the (three-fold) synthesis is “association” (Kant, A 123). And, the law governing the unity of the synthesis, i.e., the reference through judgments to apperception, is “non-contradiction” (Kant, A 151/B 191), so it is here that one should speak of an inferential norm-governed space of reasons revealed through rational synthetic processing.

Now, most importantly, the synopsis hangs together by the law of affinity, and the initial “fold” of the three-fold synthesis must “run through” the synopsis in a certain way. “For apprehension is only a placing together of the manifold of empirical intuition; and we can find in it no representation of any necessity which determines the appearances thus combined to have connected existence in space and time” (Kant 1998, 298; A 177/B 219). Kant’s point that no necessity determines the appearances is not to say that an inferential necessity is grounded in the function of norms. Rather, it is to say that affinity non-discursively differentiates an agent’s external environment such that the subsequent “connected”-ness determines a state of affairs. Hence, the determination of a state of affairs is habit-governed, not norm-governed.

Blurring the distinction between receptivity and spontaneity in commenting on Kant’s Critique of Pure Reason, Brandom makes explicit the mistake of considering the “synopsis” to be itself a “synthesis.” Of course, according to Kant, the synopsis is not a synthesis. For example, because Brandom refers to the synopsis as a synthesis, “there is synthesis in intuition and imagination also [my emphasis]” by way of chain argument, i.e., since synthesis occurs in both sensibility and understanding, and “synthesizing activity is an aspect of judgment,” he arrives at the mistaken conclusion — central to his and the Pittsburgh School’s project — “Thus all our cognitive activity consists of judgment and aspects of that activity” (Brandom 1994, 80). Whereas, in the language I used above, Brandom sees a state of affairs as formed through the synthesis of implicit conceptual, i.e., discursive, theses regarding an agent’s experience of the external environment, he should have said that a state of affairs is a primordially memory-governed contingent formation of an agent’s potential relations to their synoptic Umwelt, i.e., the totality of reciprocally related potential instantiations limited by an agent’s perceptual capacities.

Notice, Brandom’s above reading of synopsis in Kant as synthesis is a precursor for his later claim that “In an important sense there is no such boundary [between the conceptual and the nonconceptual], and so nothing outside the realm of the conceptual” (Brandom 2000, 357). And, Brandom’s footnote to this quote states, “Although the details are very different, at this level of abstraction there is on this point deep agreement between the views expressed in Making It Explicit and those John McDowell puts forward in his important, powerful, and original Mind and World” (Brandom 2000, 357). Hence, I read the Pittsburgh School generally, and specifically Brandom’s belief that he has provided proof of discursivity throughout perceptual experience, to derive from a mistaken reading of the Transcendental Deduction in Kant’s Critique of Pure Reason (cf. Scalambrino, 2011).

So, when central to NF Brandom says, “conceptual contents might be attached to states and performances functionally, in virtue of their relations to one another and to elements of the environment to which they apply” (Brandom 2009,177-8), he pulls in the assumption of an independent environment filled with objects that answer to the normative space of reasons. On the one hand, what is implicit in this assumption is the conclusion Brandom will later make explicit, i.e., metaphysical realism. On the other hand, I have already shown that this mathematical image of the external world seems to derive from (a) collapsing the discovery/demonstration distinction, (b) not affirming the differentiating function of non-discursive habit toward determining the state of affairs for perceptual experience, and (c) an interpretation of the Transcendental Deduction in the Critique of Pure Reason that seems inconsistent with Kant’s Copernican revolution.

What remains, then, is to respond to two questions regarding tautology and post hoc inference: (1) to what does counter-factual reasoning pertain if the position of NF excludes awareness that the state of affairs to which an experiential hypothesis relates is contingent upon an agent’s habit-governed relation to its external environment; (2) when NF halts the regress, to what does the regress pertain?

First, for the normative functionalist, counter-factual reasoning is supposed to provide support for NF by showing that, for example, if the door had not been open it would have been closed. Yet, as I hope is clear at this point in the article, this counter-factuality relates to an already agreed upon state of affairs. So, to ask within the demonstrable constraints of valid inference if there were another logical possibility than open or closed is to have already shifted from the dynamic to the mathematical, i.e., from discovery to demonstration. Moreover, to use counter-factual reasoning regarding perceptual experiences that did not occur is to post hoc postulate within a state of affairs that need not have been the state of affairs that would have been perceived. To be clear, I do not consider Kant, the standard criticism of NF, or myself to be attempting an argument against the rules of valid inference. Rather, the question is how can inferential reasoning be correct regarding perceptual experience while at the same time not revealing the mind external world as-it-is-in-itself? Kant, of course, already answered this question.

Second, the logic of the regress itself pertains to the dimension of demonstration. That this is so can be seen by noticing that the discursive identification of an environmental perception receives its determination within a norm-governed inferentially constellated space of reasons. And, since the dimension of demonstration is norm-governed, of course what halts the regress does not rely on a (The Myth of the) Given foundation; what halts the regress will ultimately be some norm arrived at by traversing the inferential logic constellating the space of reasons from which the environmental perception in question received its meaningful identity.

That the normative functionalist position affirms a tautology is only a problem in so far as its tautological borders are construed too widely. To say that the norm-governed space in which assertions have meaning is the space that halts logical regress is ultimately to state a tautology. Recall Maher, “norms or rules of reasoning are important not simply because one must be responsive to them in order to think or speak at all, but also because they constitute the very meaning or content of our thoughts and claims” (Maher 2012a, 5). Just as A = A: the inferentially constellated normative space on which we base the meaning of our identifications is the space in which we give and ask for reasons so as to demonstrate the appropriateness of our identifications, and to give and ask for reasons is to provide a basis for the meaning of our identifications. Yet, normative functionalists go too far by not acknowledging the non-norm-governed functions contributing to experience.

V. Conclusion

In this article, I have shown that normative functionalists reduce the differentiating function of non-discursive habit in experience to post hoc logical possibility. This discussion helps fill-in the Other Pittsburgh School’s critical position, since it shows how the Pittsburgh School commits itself to the position of metaphysical realism, i.e., the claim to know mind independent external reality. On the one hand, habit influences perceptual experience by determining the state of affairs in which instantiated objects of experience are conceptually identified. On the other hand, taking normativity as a point of departure, normative functionalists account for mind independent external reality by considering the inferential relations that pertain to the norm-governed identification of the object instantiated in perceptual experience. Yet, such accounts fail to consider the states of affairs non-inferentially related to the norm-governed identification of the instantiated object of experience. And, these states of affairs would need to be included in an accounting for mind independent reality. Hence, the Pittsburgh School’s understanding of perception as norm-governed misunderstands the role of habit in perception. As a result, the Pittsburgh School should reduce the domain of experience to which the position of NF is said to pertain.

Further, I have specifically shown that the Pittsburgh School’s construal of perception is too wide. This overly wide construal of perception seems to mislead normative functionalists to include an affirmation of metaphysical realism as part of the position of NF. I advocated for a narrowed construal of norm-governed-perception by discussing non-norm-governed, i.e., habit-governed, aspects of perceptual experience. More specifically, the position of NF does not consider the function of procedural memory involving perceptual relations, to differentiate which aspects of the environment are instantiated into lived perceptual experience. I, then, showed that the version of NF corrected of its overly wide construal of perception is tautologically grounded. We do not have knowledge of the world until our claims about the world are justified. Whereas it is this tautologically grounded inferentially constellated norm-governed space of reasons that halts the regress as anti-foundational and is operable in regard to counter-factual reasoning, consistent with Kant’s Copernican revolution the space of reasons cannot provide knowledge of mind independent external reality. Hence, I have advocated for the incorporation of perceptual memory into the normative functionalist research project as a bulwark against tautological arguments that might otherwise lead astray into metaphysical realism.


Aristotle. 2004. Nicomachean Ethics. Translated by Roger Crisp. Cambridge: Cambridge University Press.

––––––. 1962. Prior Analytics. Translated by Hugh Tredennick. London: William Heinemann.

Brandom, Robert. 1994. Making It Explicit: Reasoning, Representing, and Discursive Commitment. Cambridge, MA: Harvard University Press.

––––––. 2009. Reason in Philosophy: Animating Ideas. Cambridge, Mass: The Belknap Press of Harvard University Press.

––––––. 1979. “Freedom and Constraint by Norms.” American Philosophical Quarterly 16 (3): 187-196.

––––––. 2000. “Facts, Norms, and Normative Facts: A Reply to Habermas.” European Journal of Philosophy 8 (3): 356-374.

Craik, Fergus I.M. and Bialystok, Ellen. (2006). “Cognition through the lifespan: mechanisms of change,” Trends in Cognitive Sciences 10 (3): 131-138.
deVries, Willem. Wilfrid Sellars. New York: McGill-Queen’s University Press.

Erickson, Michael A. 2008. “Executive attention and task switching in category learning: Evidence for stimulus-dependent representation,” Memory & Cognition 36 (4): 749-761.

Habermas, Jürgen. 2003. “From Kant to Hegel: On Robert Brandom’s Pragmatic Philosophy of Language.” In Truth and Justification, edited by Barbara Fultner, 131-173. Cambridge, MA: MIT Press.

Huys, Raoul, Nicholas J Smeeton, Nicola J Hodges, Peter J Beek, and Mark A Williams. 2008. “On the dynamic information underlying visual anticipation skill.” Perception & Psychophysics 70 (7): 1217-1234.

Kant, Immanuel. 1998. Critique of Pure Reason. Translated by Paul Guyer and Allen Wood. Cambridge: Cambridge University Press.

Kuhn, Thomas S. 1962. The Structure of Scientific Revolutions. Chicago: University of Chicago.

Maher, Chauncey. 2012a. The Pittsburgh School of Philosophy. New York: Routledge.

––––––. 2012b. “Reply to Reider.” Social Epistemology Review and Reply Collective 1 (11): 16-23.

Margolis, Joseph. 2007. “Present doldrums, pleasant prospects: Philosophy early in the new century.” Philosophy and Social Criticism 33 (1): 15-34.

McDowell, John. 1994. Mind and World. Cambridge, MA: Harvard University Press.

Micheyl, Christophe, Josh H McDermott, Andrew J. Oxenham. 2009. “Brief Reports: Sensory noise explains auditory frequency discrimination learning induced by training with identical stimuli.” Attention, Perception, & Psychophysics 71 (1): 5-7.

Pascal, Blaise. 1995. Pensées. Translated by A.J. Krailsheimer. London: Penguin Classics.

Proust, Marcel. 2003. In Search of Lost Time, Vol. I: Swann’s Way. Edited by D.J. Enright. Translated by C.K. Scott Moncrieff and Terence Kilmartin. New York: The Modern Library.

Reider, Patrick J. 2012. “Pittsburgh and the analytic tradition in philosophy.” Social Epistemology Review and Reply Collective 1 (9): 20-27.

Rockmore, Tom. 2010. “Some Recent Realist Readings of Hegel.” In Hegel and the Analytic Tradition edited by Angelica Nuzzo, 158-172. London: Continuum.

Rorty, Richard. 1979. Philosophy and the Mirror of Nature. New Jersey: Princeton University Press.

Scalambrino, Frank. 2011. “Non-Being & Memory: A critique of pure difference in Derrida and Deleuze.” Doctoral dissertation, Duquesne University. Retrieved from ProQuest. (UMI: 3466382).

––––––. 2012. “Mnemo-psychography: The Origin of Mind and the Problem of Biological Memory Storage.” In Origins of Mind edited by Liz Swan, 327-339 Biosemiotics, Vol. 8. New York: Springer.

Sellars, Wilfrid. 1956. Empiricism and the Philosophy of Mind. Minnesota Studies in the Philosophy of Science, Vol. I, edited by Herbert Feigl and Michael Scriven, 253-329. Minneapolis, MN: University of Minnesota Press.

Smilkstein, Rita. 2003. We’re Born to Learn. London: Sage Publications.

Sperling, George. 1960. “The information available in brief visual presentations.” Psychological Monographs 74 (1): 1-29.

Swindal, James. 2007. “Can a discursive pragmatism guarantee objectivity?: Habermas and Brandom on the correctness of norms.” Philosophy and Social Criticism 33 (1): 113-126.

Wolf, Kerstin, Daniel Ebeling and Notger G Müller. 2009. “The effects of implicit attentional learning and habituation on inhibition of return.” Attention, Perception, & Psychophysics 71 (1): 26-41.

Wittgenstein, Ludwig. 1922. Tractatus Logico-Philosophicus. Translated by Charles Kay Ogden. New York: Harcourt, Brace & Co.

Wittgenstein, Ludwig. 1969. On Certainty. Translated by Denis Paul and G.E.M. Anscombe, Oxford: Basil Blackwell.

Wurzer, Wilhelm. 1992. “Nietzsche and the Problem of Ground.” In Anti-foundationalism Old and New edited by Tom Rockmore and Beth Singer, 127-142. Philadelphia: Temple University.

[1] Brandom, Robert. 1979. “Freedom and Constraint by Norms.” American Philosophical Quarterly 16 (3): 187-196. Page 192.

[2] Cf. Habermas, p. 140; though Habermas and Margolis are not necessarily affiliated with Duquesne University, the rest of the thinkers on the list are, thereby demonstrating the appropriateness of associating this position with Duquesne.

[3] On the importance of Kant for the Pittsburgh School see Rorty, 1979, p. 149, and Brandom, 2009, p. 114.

Categories: Articles

Tags: , , , , , , ,

1 reply


  1. Employees as Sims? The Conflict Between Dignity and Efficiency, Frank Scalambrino « Social Epistemology Review and Reply Collective

Leave a Reply