“What” by Inferential Semantics

Brandom’s inferential semantics can be seen as providing a general framework for answering “what is…” questions. Semantics is about meaning — especially of concrete things said — and inferential semantics is about understanding meaning as a kind of practical doing involved with reasons. Looked at this way, a meaning reflects an inferential role, or role in real-world reasoning. Such roles always have two sides — conditions for appropriate use, and consequences of using this rather than that. Brandom identifies conceptual content with such inferential roles, and focuses on a contrast between these and simple definition, but I want to emphasize instead that all simple definition should be understood as a kind of summary of what implicitly distinguishes a particular inferential role from others.

The kind of meaning of interest here is in principle shareable rather than subjective, private, or psychological. Meaning is social and essentially involved with communication, but it is not a matter of empirical fact. Rather than explaining communication in terms of empirical facts, we should ultimately explain what we call empirical facts in terms of well-founded shareable meaning. The more we are able to explicitly spell out conditions of use and consequences of things that are said, the more substantive content we can share with others.

The “what is…” questions classically asked by Plato and Aristotle have an open-ended character because they are concerned with what something means for a reasoning being in general, which is an open-ended context. To have meaning for a reasoning being is to make a difference in the way the being reasons in life. In this way, Plato and Aristotle also were deeply concerned with the inferential roles of things, and practiced a kind of inferential semantics. This is ultimately inseparable from questions of goodness of reasoning. Here, too, inferential semantics depends on normative pragmatics.

Kant and Foundationalism

According to Kant, all human experience minimally involves the use of empirical concepts. We don’t have access to anything like the raw sense data posited by many early 20th century logical empiricists, and it would not be of much use if we did. In Kantian terms, this would be a form of intuition without concepts, which he famously characterized as necessarily blind, and unable to function on its own.

Foundationalism is the notion that there is certain knowledge that does not depend on any inference. This implies that it somehow comes to us ready-made. But for Kant, all use of empirical concepts involves a kind of synthesis that could not work without low-level inference, so this is impossible.

The idea that any knowledge could come to us ready-made involves what Kant called dogmatism. According to Kant, this should have no place in philosophy. Actual knowledge necessarily is a product of actual work, though some of that work is normally implicit or preconscious. (See also Kantian Discipline; Interpretation; Inferentialism vs Mentalism.)

It also seems to me that foundationalism is incompatible with the Kantian autonomy of reason.

Historiography, Inferentialism

Having laid out some preliminaries, I’ve begun to circle back to more questions of historical detail related to the development here, and it seems fitting to summarize the motivations driving these more historical notes. History is all about the details, but in any inquiry, what are actually higher-order questions about methodology ought to inform primary investigations. We never just have data; it always has to be interpreted, and this involves questions about methodology. With history, this often involves critical examination of the applicability of categories that may tend to be taken for granted. Thus, I am adding notes about the application of various categories or concepts in particular historical settings, and about historical details that seem to have larger methodological significance.

I’m looking back at the history of philosophy (and, to some extent, broader cultural developments) from a point of view inspired by the “inferentialism” of Brandom (taking this as a general name for his point of view), as well as by my own ideas for a revitalized Aristotelianism. In Tales of the Mighty Dead and elsewhere, Brandom himself has effectively placed the historical roots of his development in the broad tradition of early modern philosophical rationalism, including the work of Descartes, Spinoza, and Leibniz. I find standard connotations of the term “rationalism” rather problematic, and want to separate Descartes — of whom I am much more sharply critical than Brandom seems to be — from Spinoza and Leibniz, for whom I find additional reasons to be sympathetic. Brandom has contributed to a new understanding of Kant, and has developed a landmark reading of Hegel. I want to help support the broad thrust of these with historical considerations, while reconnecting them with fresh readings of Aristotle, Plato, and other historical philosophers. With some caveats and in spite of Brandom’s own brief comments, I also want to suggest a possible rapprochement with key insights of 20th century French “structuralism”.

A key point common to most of the tendencies mentioned above is an emphasis on the role of difference in making things intelligible. In the context of philosophical arguments, this means that critical distinctions are as important as positive assertions. Contrasts not only greatly facilitate but largely shape understanding. Brandom himself has developed the contrast between inferentialism and the representationalism of Descartes and Locke. He has made large use of Wilfrid Sellars’ critique of a “Myth of the Given” associated with most varieties of empiricism, and has also referenced the critique of psychologism developed by Frege and others in a logical context.

I have been using the term “mentalism” for a privileging of contents that are supposed to be immediately present to a personal “mind” that is itself conceived mainly in terms of immediate awareness. It seems to me that Descartes and Locke’s version of this was a historically specific combination of all the above notions from which an inferentialism would seek to distinguish itself — representationalism, the Myth of the Given, and psychologism. I have been concerned to point out not only that Cartesian-Lockean mentalism has historically specific antecedents that long predate modernity (going back to Augustine, with some foreshadowing in Plotinus), but also that a proto-inferentialist countertrend is actually even older, going back to Plato and Aristotle’s emphasis on the primacy of reason and reasoned development.

In A Spirit of Trust, Brandom has among many other things expanded on Hegel’s critique of Mastery. I find this to be of tremendous importance for ethics, and consonant with my structuralist sympathies. I have been concerned to point out how extreme claims of mastery are implicit in the various historical kinds of voluntarism, which all want to put some notion of arbitrary will — or authority attributed one-sidedly to such a will — ahead of consideration of what is reasonable and good.

Usual generalization caveats apply to statements about “isms”. In any particular case where the terms seem to apply, we need to look at relevant details, and be alert to the possibility that all aspects of a generalized argument may not apply straightforwardly. (See also Historiography; History of Philosophy.)

Inferentialism vs Mentalism

Brandom’s “inferentialism” or emphasis on material inference effectively makes what I call ethical reason the most important thing in the constitution of subjectivity — not psychology, and not some putative immediate mental presence, or universal transparent representational medium, or supposedly perfect reflexivity.

This is not to deny that there is such a thing as immediacy; it is rather to specify that immediacy is not foundational, and has nothing to do with certainty. Immediacy has a very different role to play, in showing us the world’s “stubborn recalcitrance to mastery and agency” and providing occasions for learning. (See also Mind Without Mentalism; Psyche, Subjectivity.)

Empiricism

Already in the 1950s, analytic philosophers began to seriously question empiricism. Quine’s “Two Dogmas of Empiricism” (1951), Wittgenstein’s Philosophical Investigations (1954), and Sellars’ “Empiricism and the Philosophy of Mind” (1956) all contributed to this.

Brandom explicates Sellars’ pivotal critique of the empiricist “Myth of the Given” as belief in a kind of awareness that counts as a kind of knowledge but does not involve any concepts. (If knowledge is distinguished by the ability to explain, as Aristotle suggested, then any claim to knowledge without concepts is incoherent out of the starting gate.) Building on Sellars’ work, Brandom’s Making It Explicit (1994) finally offered a full-fledged inferentialist alternative. He has rounded this out with a magisterial new reading of Hegel.

The terms “empiricism” and “rationalism” originally referred to schools of Greek medicine, not philosophy. The original empirical school denied the relevance of theory altogether, arguing that medical practice should be based exclusively on observation and experience.

Locke famously began his Essay Concerning Human Understanding (1689) with an argument that there are no innate ideas. I take him to have successfully established this. Unfortunately, he goes on to argue that what are in effect already contentful “ideas” become immediately present to us in sensible intuition. This founding move of British empiricism seems naive compared to what I take Aristotle to have meant. At any rate, I take it to have been decisively refuted by Kant in the Critique of Pure Reason (1781; 2nd ed. 1787). Experience in Kant is highly mediated. “Intuitions without concepts are blind.” (See also Ricoeur on Locke on Personal Identity; Psyche, Subjectivity.)

In the early 20th century, however, there was a great flourishing of phenomenalism, or the view that all knowledge is strictly reducible to sensation understood as immediate awareness. Kant himself was often read as an inconsistent phenomenalist who should be corrected in the direction of consistent phenomenalism. Logical empiricism was a diverse movement with many interesting developments, but sense data theories were widely accepted. Broadly speaking, sense data were supposed to be mind-dependent things of which we are directly aware in perception, and that have the properties they appear to have in perception. They were a recognizable descendent of Cartesian incorrigible appearances and Lockean sensible intuition. (Brandom points out that sense data theory is only one of many varieties of the Myth of the Given; it seems to me that Husserlian phenomenology and its derivatives form another family of examples.)

Quine, Wittgenstein, and Sellars each pointed out serious issues with this sort of empiricism or phenomenalism. Brandom’s colleague John McDowell in Mind and World (1994) defended a very different sort of empiricism that seems to be a kind of conceptually articulated realism. In fact, there is nothing about the practice of empirical science that demands a thin, phenomenalist theory of knowledge. A thicker, more genuinely Kantian notion of experience as always-already conceptual and thus inseparable from thought actually works better anyway.

Thought and intuition are as hylomorphically inseparable in real instances of Kantian experience as form and matter are in Aristotle. A positive role for Kantian intuition as providing neither knowledge nor understanding, but crucial instances for the recognition of error leading to the improvement of understanding, is preserved in Brandom’s A Spirit of Trust. (See also Radical Empiricism?; Primacy of Perception?; Aristotle, Empiricist?)

Constructive

Brandom’s inferentialist alternative to representationalism stresses material, meaning-oriented over formal, syntactic inference. Prior to the development of mathematical logic, philosophers typically used a mixture of reasoning about meanings with natural language analogues of simple formal reasoning. People in ordinary life still do this.

Where Brandom’s approach is distinctive is in its unprecedentedly thorough commitment to the reciprocal determination of meaning and inference. We don’t just do inference based on meanings grasped ready-to-hand as well as syntactic cues to argument structure, but simultaneously question and explicitate those very meanings, by bracketing what is ready-to-hand, and instead working out recursive material-inferential expansions of what would really be meant by application of the inferential proprieties in question.

For Brandom, the question of which logic to use in this explicitation does not really arise, because the astounding multiplication of logics — each with different expressive resources — is all in the formal domain. It is nonetheless important to note that formal logics vary profoundly in the degrees of support they offer for broad representationalist or inferentialist commitments.

Michael Dummet in The Logical Basis of Metaphysics argued strongly for the importance of constructive varieties of formal logic for philosophy. Constructive logics are inherently inference-centered, because construction basically just is a form of inference. (Dummet is concerned to reject varieties of realism that I would call naive, but seems to believe the taxonomy of realisms is exhausted at this point. This leads him to advocate a form of anti-realism. His book is part of a rather polarized debate in recent decades about realism and anti-realism. I see significant overlap between non-naive realisms and nonsubjective idealisms, so I would want to weaken his strong anti-realist conclusions, and I think Brandom helps us to do that.)

Without endorsing Dummet’s anti-realism in its strong form, I appreciate his argument for the philosophical preferability of constructive over classical logic. It seems to me that one cannot use modern “classical” formal logic without substantial representationalist assumptions, and a lot of assumed truth as well. If and when we do move into a formal domain, this becomes important.

As used in today’s computer science, constructive logic looks in some ways extremely different in its philosophical implications from Brouwer’s original presentation. Brouwer clouded the matter by mixing good mathematics with philosophical positions on intuition and subjectivity that were both questionable and not nearly as intrinsic to the mathematics as he seemed to believe. The formal parts of his argument now have a much wider audience and much greater interest than his philosophizing.

Constructive logic puts proof or evidence before truth, and eschews appeals to self-evidence. Expressive genealogy puts the material-inferential explicitation of meaning before truth, and eschews appeals to self-evidence. Both strongly emphasize justification, but one is concerned with proof, the other with well-founded interpretation. Each has its place, and they fit well together.

Syntax, Semantics, Ethics

How we understand one another in our social interactions is of paramount importance for ethics. Pointing, gestures, and similar cues can get us started, but we cannot get very far without considering linguistic meaning, and we cannot get very far with natural language meaning without considering implicit and explicit inferences.

In concrete utterances, syntax still plays a large role in specifying the overall shape of the inferences a speaker is implicitly asking us to endorse with respect to some content in question. Here we are concerned not with formal definition of syntactic features, but with specific, concrete usage that we implicitly, defeasibly take as specifying definite higher-level inferential connections by virtue of the grammar employed.

By understanding the structure of a speaker’s overall argument from syntactic as well as semantic cues, we get all sorts of nuances like intended qualifications and specifications of scope that can be all-important in assessing the reasonableness of what is being said. How well we do this depends at least partly on us as well as what was said, and also on our familiarity with the speaker’s particular speech patterns, which may vary from what is common or usual. We can also silently compare the speaker’s speech patterns to what is common or usual, and wonder if what they seemed to say was what they actually meant; or kindly point out to them that what they actually said could be misunderstood. (See also Inferential Semantics; Honesty, Kindness.)

Plato and Aristotle Were Inferentialists

In the context of modern philosophy, Brandom has developed an important contrast between representationalism and inferentialism. Representationalism says that representation comes before inference in the order of explanation, and inferentialism says that inference comes first.

Plato was very pessimistic about the potential of representation, as witnessed by the dialogues’ discussions of “imitation”, and the treatment of writing in Phaedrus. By contrast, inference or reasoning is presented as the main way to truth in the dialogues. Inference — and not representation — is what is primarily appealed to in the validation or invalidation of assertions. (See also Dialogue; Platonic Truth.)

Aristotle was less pessimistic about representation, but even more concerned with inference. He was the great originator of the world’s first developed logic, which was in fact centered on inference rather than truth values. While taking pioneering steps toward formalization, he also devoted much attention to definition, meanings of terms, and their distinctions and ambiguities in concrete usage (see Aristotelian Semantics; Aristotelian Demonstration). Aristotle distinguishes between inference based on the fact, which is a kind of formal inference, and inference based on the meaning, which is the material inference of Sellars and Brandom, also known to medieval logicians. Further, Aristotle’s elementary criteria for truth and falsity depend on material inference (see Aristotelian Propositions).

The kind of representation Brandom is particularly concerned with, which he attributes to Descartes, is based on isomorphism rather than resemblance. As an aside, I tend to think there was a notion of isomorphism in the ancient world, though it is a little hard to separate from resemblance. Euclid talked about similar triangles, which are technically an example of both. Aristotle would certainly say that resemblance is “said in many ways”, one of which could be isomorphism. I think given the opportunity he would say, for instance, that individual concretely uttered words are at some level isomorphic to whatever meanings those words turn out to have in some context. The words do not resemble their meanings. (See also Historiography, Inferentialism; Inferentialism vs Mentalism;.)