Andy Blunden September 2011
Given the failure of analytical science to even provide itself with an adequate idea of what a concept is, let alone elaborate a systematic psychology of concepts, Robert Brandom’s philosophical study of concepts, is all the more to be welcomed because Brandom situates himself squarely within the tradition of analytical philosophy, which is also his main protagonist.
A former student of Richard Rorty, Brandom identifies himself both as a Pragmatist and as an analytic philosopher in the “Anglo-American tradition.” His pragmatic reading of Kant is particularly valuable, but his attempt to extend this approach to Hegel I find less successful. Brandom’s approach is like that of Robert R. Williams in which Hegel is cast as a liberal with a philosophy of unmediated interactions. Mediation is the sine qua non of Hegel philosophy.
His 2000 book, Articulating Reasons, opens with the words: “This is a book about the use and content of concepts.” He claims:
I am putting forward a view that is opposed to many ... of the large theoretical, explanatory, and strategic commitments that have shaped and motivated Anglo-American philosophy in the twentieth century: empiricism, naturalism, representationalism, semantic atomism, formalism about logic, and instrumentalism about the norms of practical rationality, [but] I take my expository and argumentative structure and the criteria of adequacy for having made a claim with a clear content, argued for it, and responsibly followed out its consequences resolutely from the Anglo-American tradition.
In particular, unlike the rest of the analytical tradition, Brandom is interested in what is distinctive about concept-use, rather than taking concept-use as simply a more developed form of the behaviour of a trained pigeon. Consequently, he is more interested in how concepts function as premises for reasoning, rather than just as criteria for classification. Understanding, after all, means knowing what is entailed by a concept, not simply differentially responding to this or that condition in the environment, something that even machines and the lower animals can do. Brandom describes his approach as “inferentialism” in contrast to the dominant view of concepts in analytical philosophy which he calls “representationalism,” i.e., taking concepts to simply be representations of their object.
One of the features of Brandom’s approach, which he credits to Kant, is that without obliterating the distinction between is and ought, he takes as his topic normativity without distinguishing between norms flowing from belief (theoretical norms), norms of desire (practical or ethical norms) and norms of meaning (semantic norms). Indeed, it would seem that there is no hard line to be drawn between adhering to a norm enforced by social sanctions, one enforced by the laws of nature or by the shared understandings of a language community. This is an approach which takes the ethos and beliefs of a social formation as a whole and effectively overcomes the dichotomy between science and ethics.
Nonetheless, Brandom claims “The topic of philosophy is normativity in all its guises, and inference in all its forms” (2009: 126), and Brandom almost never ventures outside the domain of philosophical enquiry so-defined. Since norms are, by definition, generally known and can be taken as given, he eschews interest in the history, sociology or psychology of norms. I think the days when a thinker can usefully contribute to an understanding of the human condition without reference to the world outside of Logic were over some time in the 1830s. Consequently, the answer Brandom gives to the question “What is a concept?” is an answer belonging solely to Logic, leaving unanswered all the questions of Ontology, Social Theory and Psychology, which bedevil the problem of explaining what a concept is.
Brandom sets out from the beginning to identify what is distinctive about concepts, that is to say, about the use of concepts and the creatures, human beings, who use concepts. It may be that there is no firm line to be drawn in a spectrum which goes from a thermostat which turns on the cooler when the temperature exceeds its set point, to the physicist who grasps the concept of critical mass. But there is a difference, and it is this difference which is important.
The differential response of a machine or in fact any inanimate object, according to a stimulus, or the conditioned reflex of a pigeon trained to peck a red button to get food, indeed simulates one aspect of a concept, namely, a response which discriminates for the existence of some condition. In this same sense, a concept can be said to be a representation of its object. This capacity for representation is a universal property of matter, in that all material objects and organisms respond differentially to environmental conditions. Clearly, this does not tell us what is distinctive about concepts.
Brandom calls upon the intuition that having a concept of something means understanding it, and the thermostat may respond to temperature and cause a switch to turn on, but it cannot be said to understand, any more than the parrot understands what it means when it responds to the presentation of a red object by calling “It’s red!” The parrot does not know the significance of its words. So what then does it mean to understand?
To understand, according to Brandom, means to be able to use a concept as both a conclusion and a premise in reasoning. A concept is therefore, in the first place, not just a representation of its object. It is a predicate which could be asserted of an object, and the inferences which could be drawn from the existence of the object.
This raises a couple of points which require clarification. Firstly, the concept is not a representation of what actually exists, but rather it represents a hypothetical condition, a possible predicate of some object, or the conclusion of some reasoning process. Even sensations need to be seen as ‘evidence’ from which a representation is constructed. So Brandom’s idea of representation is quite distinct from all kinds of automatic, discriminatory responses to existing stimuli. It is not a reflection of an existing state or perceptual field, but a product of reasoning. The concept can exist without the presence of its object.
Secondly, Brandom is talking about concept-use, and his strategy is to infer what a concept is from what can be said of concept-use. As a pragmatist, he takes knowing-how as prior to knowing-that. What a concept is, is to be inferred from what can be done with it. Conscious awareness of the concept is therefore secondary to acting as if one knows the concept.
So much for concepts as ‘representations’. But understanding implies that the concept is not just a label, but means something to the subject. For example, it is simple enough to define the conditions which are formally represented by the concept of ‘AWOL’, but to understand this concept means that you know that it means arrest, danger and humiliation. These are the inferences which follow from the conditions formally represented by the concept (absent without leave). A subject cannot be said to understand the concept unless they know all these conditions which can be inferred from the concept. Note that there are various conclusions which can be inferred from the premise, AWOL, in a purely formal sense, which are contained in the concept in a formal way, and do not therefore mark the condition as a concept, properly so-called. These are what Brandom calls ‘material inferences’, which are not simply implicit in the formal conditions represented, but flow from the concept’s place in a whole network of relations (e.g. military practice), but which can be inferred by someone who truly understands the concept. To understand a concept therefore entails more than to understand the formal conditions under which the concept is extant, but in addition, to understand the whole system of concepts of which the concept is a part. That is the content of the concept, and only a human being who understands the norms of the language community in which the concept exists can make such inferences, and therefore be said to understand the concept. The concept, so to speak, channels this content from the whole system of concepts, into the particular situation of its applicability.
A couple of points of clarification are required here. Firstly this distinction between formal inference and material inference. According to Brandom it was Frege who founded the analytical current in philosophy with the publication of his Begriffsschrift in 1879, and it was Frege’s meaning that concepts entailed material inferences, and that his theory was by no means limited to formal inference. Brandom says that it was Boole who interpreted Frege in the spirit of formal inference, thereby limiting the scope of the theory to the kind of formal reasoning which is applicable only to mathematical sets, and the analytical tradition never recovered Frege’s original meaning. Brandom says that classification was the master practice which underlay Logic from mediaeval times, and it was this castration of Frege’s theory, which rendered reasoning as a process of formal categorisation by attributes and removed all material content from reasoning.
The use of the word ‘material’ to refer to what follows from a concept as a result of its interconnection with the world, rather than its formal conditions of existence, seems at first sight, a confusing choice of words. ‘Material’ seems to imply inferences which are limited to those given by natural science: thunder may be inferred from lightning, and so on. In philosophy however, the materiality of a thing means precisely the real interconnection of the thing with the rest of the universe, and marks the content of a concept as opposed to its form. As Engels put it: “The real unity of the world consists in its materiality” (MECW v.25: 41). The choice of the word ‘material’, therefore, is very apt. So material inferences equally include inferences that follow from the place of AWOL in military law, the state of military conflict at the time, the relation of the military to the surrounding population, and so on – social and psychological phenomena, just as much as natural processes. The network of concepts rests on the fundamental unity of the material world.
So at a very basic level, this is what a concept is for Brandom. It is a possible predicate of a judgment (what can be said of something, to use Aristotle’s expression), which can be the conclusion to a process of inference, and the premises for a process of inference. A concept is the source of inferences which originate in the concept’s materiality, its connection with a whole set of relations or a theory.
What kind of existence does Brandom see for concepts?
A concept is a norm of judgment. That is, concepts are norms existing in some community which determine how judgments ought to be made, whether from the point of view of the community’s metaphysical beliefs (their natural science) or their custom and practice (their ethics). As norms they are implicit in the linguistic practice and activity of a community more generally, and can be made explicit in the form of a concept. As norms of judgment, concepts are therefore the subject matter of philosophy. A concept may be applied erroneously, because norms do not determine actions in that mechanical sense, but nonetheless, a concept which is used not in accordance with norms is deemed to be used in error. So again, Brandom is a pragmatist in the sense that the meaning of a concept is to be inferred from its use.
Here Brandom comes close to Wittgenstein. He differs from Wittgenstein in that, whereas Wittgenstein does not privilege any kind of language game over any other, and says “there is no downtown in language-use,” Brandom does insist that there is a ‘downtown’, namely reasoning and inference. Concepts may figure in all kinds of activity and language games, but “applying a concept is to be understood in terms of making a claim or expressing a belief. The concept concept is not intelligible apart from the possibility of such application in judging” (2000: 160). Brandom calls himself an inferentialist, because he takes the use of concepts in reasoning to be what makes all the other language games possible.
Brandom is close to Wittgenstein in another way too, which Brandom usefully explains in terms of the units of analysis used by Kant, Wittgenstein, Frege and himself. The unit of analysis is the simplest entity which can figure in the relevant theory, and upon which all the concepts of a theory must be built. According to Brandom, Kant takes the Judgment as the unit of experience, Frege takes the smallest expression to which pragmatic force can be attached, and Wittgenstein the smallest expression whose utterance makes a move in a language game. In line with this tradition, Brandom take the proposition as his unit of analysis, so nothing smaller than a proposition can be recognised as meaningful. Brandom thus describes himself as a ‘propositionalist’. A proposition is an expression used in reasoning, often represented in symbolic logic by the letters p or q. This choice corresponds to Brandom taking reasoning as the archetypal language game, and represents Brandom’s concept of human action. A proposition is the simplest form of sentence but is slightly broader than a judgment, in that a proposition can make universal (quantifiable) claims. Brandom takes the proposition as the smallest expression to figure in his theory, his unit. He also demands that a concept must be capable of figuring as the premise or conclusion of a reasoning process. Does this mean that, contrary to normal usage, Brandom takes a concept to be a kind of proposition, rather than a subsentential expression such as a predicate or phrase? No.
By taking his unit as a proposition, Brandom remains on rigorous methodological ground. While giving us his concept of inference, whose simplest unit is a proposition, he avoids all the methodological problems which flow from trying to dissect propositions into parts with all the syntactic, pragmatic and semantic complexity of language-use, trying to define terms or expressions in some consistent way as the embodiment of a concept. Nonetheless, concept is the central object of his philosophy and concepts are associated with subpropositional expressions, such as predicates. Brandom resolves this by holding that concepts are to be inferred from their use in propositions. The fundamental form of the conceptual is the propositional, and the concept is to be inferred from the proposition. Concepts are the norms or rules for forming judgments, and can be inferred from the use of words in propositions. A concept is a predicate of a possible judgment.
Brandom makes all this very clear with his explanation of bad concepts. These are concepts which we don’t use because their very utterance carries the inference that certain propositions follow from certain conditions, and if we don’t agree with this inference we simply must not use the word. The clearest examples are terms reflective of racial prejudice such as ‘nigger’, which applies to Afro-Americans and carries the inference that it is a contemptible person. Such a term simply cannot be uttered. Another example Brandom gives is the concept of ‘blasphemy’, of which the accused Oscar Wilde said “is not one of my words.” These ‘bad concepts’ demonstrate that concepts have real content, specifically that concepts embody both the state of affairs they describe and the meaning or significance which could be ascribed to that state of affairs.
Given that concepts have real content in this sense, we can see that the use of concepts commits any person who uses a concept to the work of integrating concepts into a single whole, which is a person’s world view; a person must answer for what flows from the concepts they use. When a rational person is presented with a new concept, its ramifications and its interaction with all the other concepts must be worked through. Incompatible concepts cannot be carried side by side with each other. Thus Brandom gives us an approach to understanding the development of the rational person. The same observations apply to the development of a science. “What makes it a unified whole is the rational relation among its parts” (2009: 52). Note that there is no implication here of any kind of ‘master principle’. According to Brandom’s pragmatism, the unified whole is only the outcome of the integrative work of a rational person.
Brandom also uses this idea to represent the intelligibility of the process of historical development, following the conception of the evolution of law worked out by the early American Pragmatist, Oliver Wendell Holmes Jr.. In this idea, a new precedent is set by a judge in the light of previous decisions, rather than by reference to general principles. In setting the new precedent, the judge takes previous decisions into account and acts consistently with them, but he or she is not obliged to deduce or justify the decision in terms of the precedents in the manner of a formal logical theorem. Brandom takes this pragmatist conception of Reason as a model to represent the process of the unfolding of history. He takes it as an alternative to Hegel’s conception of the intelligibility of history, in which the concept pre-exists its manifestation in history. Holmes had put it this way: “It is the merit of the common law that it decides the case first and determines the principle afterwards” (quoted in Menand 2001: 338). At the basis of intellectual life must lie norms that are simply implicit in our practices. Thus, his principal project is to make ‘explicit the implicit structure characteristic of discursive practice as such” (Brandom 1994: 649). And it is not only language-use in which concepts are implicit, but the entirety of material culture, the use of which is a key part of the process of acquiring the concepts of our culture.
One of the problems which arises in the study of concepts is how to resolve the conflict between several theories of the nature and origin of concepts, each of which seem to have some merit. These theories are (1) Empiricism, which sees the origin of concepts in experience, (2) Pragmatism, which sees the origin of concepts in their significance for action, and (3) Rationalism, which sees concepts in terms of their capacity for the production of good inferences in reasoning. Although Brandom sees himself as a rationalist and takes the rationalist view to be the decisive one, he suggests an eclectic approach, which recognises that all these sources play a role in the formation of concepts, and in their use in different circumstances.
So Brandom takes a concept to be a property of a proposition, for all intents and purposes a subpropositional unit. But concepts gain their content not from the proposition, but from its interconnection with other concepts, so meaning arises from this whole network of concepts. Brandom poses the problem this way:
So the inferential significance of a belief depends on what else one believes. Thus the unit of meaning should be taken to be a whole theory, not just a single sentence (2000: 167).
Brandom presents a damning critique of the Psychology of Concepts as developed by cognitive psychology, a branch of science connected with the Anglo-American tradition of analytical philosophy of which he is a part. It is of particular value that Brandom’s critique is internal to analytical philosophy.
The central charge directed against the Psychology of Concepts is that their concept of concept is restricted to Representation. Jerry Fodor strenuously defends what he calls a “representational theory of mind,” providing the philosophical justification for psychological research based on this conception. Representations may be more or less complex, but at root, representing something simply means responding differentially to features in the environment.
Brandom points out that a differential response to stimuli is far from what is distinctive about concepts. But this is exactly what the Psychology of Concepts takes as its object. Even a piece of iron rusts differentially according to the presence of water and oxygen in its environment, and in that sense makes a representation of an aspect of its environment. All the research on concepts by cognitive psychology has focused on disclosing how the mind represents objects, and no attention has been paid to how a subject understands the significance of a concept, what is entailed by the concept. By focusing only on a function which human beings share with inanimate objects, machines and the lower animals, cognitive psychology has failed to shed any light on what is distinctive about concept-use, or for that matter, shown any recognition that such a distinction exists.
Granted that human beings form representations of objects in their environment, but this sheds little light on how a subject understands the object if at all, and indeed, no effort has been made to clarify what it might mean to ‘understand’ a concept. That a subject forms some kind of representation of an object or classifies the object under some category, still tells us nothing about how the representation enters into a reasoning process or has significance for action. A camera or a voice-recorder can make a representation, but is for that no closer to using the representation in a reasoning process. Drawing on Wittgenstein, Brandom points out that there is no conceivable kind of language game which could be played exclusively with representations. A concept which contributes nothing to the activity of reason or to discourse can surely not warrant the name of ‘concept’.
Concepts do not only label their objects, but also describe the object. The only sense in which Cognitive Psychology addresses this aspect of concepts is its subordination to the practice of classification, of ordering objects into sets and subsets, a practice which stretches back to the Scholastics, underpinning traditional syllogistic logic, and represents an extremely restricted type of judgment. The only effort that Cognitive Psychology makes to address complex concepts is to allow for the union, intersection and negation of sets defined by bundles of attributes.
This is the drift of Brandom’s critique of “representationalism” in contrast to his own “inferentialism.” According to Brandom, the insights which are lacking in cognitive science were there in 1879, when Frege founded analytical philosophy. Is it Boole who is responsible for analytical science having so lost its way?
We analytic philosophers have signally failed our colleagues in cognitive science. We have done that by not sharing central lessons about the nature of concepts, concept use, and conceptual content that have been entrusted to our care and feeding for more than a century (2009: 197).
And reflecting on the hierarchy of concepts which can be constructed on the basis of his own logical investigation of the formation of concepts, and speculating how this hierarchy might be reflected in the phylogenetic and ontogenetic development of concept use, he reflects that:
These are merely examples of potentially important questions raised by the hierarchy of conceptual complexity that cognitive scientists have by and large not been moved to so much as to ask. Why not? I think it is pretty clear that the answer is ignorance (2009: 223).
This criticism demonstrates that a number of serious deficiencies in the Psychology of Concepts and analytical science and philosophy generally, despite being characteristic of the analytical tradition, may not be necessary and essential features of analytical science. If we accept Brandom’s internal critique of analytical science, then a critique of Brandom’s theory of concepts perhaps offers a more significant critique of the analytical approach to the study of concepts.
Brandom identifies himself as a propositionalist. This means that a proposition is the smallest unit which he takes to contain all the essential properties of intellectual life. Concepts are taken to be properties of propositions. But it turns out that the content of the concept derives not from within the proposition, but arises through the interconnection of the concept with a whole network of inferential relations with other concepts. Brandom himself observes: “Thus the unit of meaning should be taken to be a whole theory, not just a single sentence” (2000: 167). But Brandom has not taken this step, and indeed, he is not theoretically equipped to take this step. The concept of a “whole theory” lies outside the scope of his philosophy, because in line with the Pragmatist tradition he has taken individual actions or interactions as the ultimate reality. This is essentially the same position Brandom takes when he seeks to render Hegel as a philosopher of Recognition, taking the unmediated interactions between two individuals as the ultimate reality and unit of analysis. This makes history look like a game of billiards with nothing but one-on-one interactions on a perpetually level playing field.
The metaphor of judge-made law cited above, which is a pragmatic rendering of Hegel’s conception of sprit, by disposing of the need for a pre-existing principle governing the development of new propositions, seems to justify the idea that the whole process of cultural and historical development can be rendered as interactions between individuals. But this does not stand up. The process depends essentially on the availability of the precedents, the body of enacted law and all the legal principles which exist in the form of documents. These documents are crucial mediating artefacts which regulate the development of the common law. The idea that the judge is able to make explicit what was merely implicit in the previous decisions is an attractive and eminently Hegelian idea. But it presupposes that these documented decisions act as mediating elements in the development of law, not to mention the entire material culture which supports the way of life in which the decisions are made by judges and enforced by a state.
A proposition appears to be something created and enacted in the moment when two people interact, but neither the language used in the interaction nor the concepts which are embedded in the language are created de novo in that interaction. The words and concepts relied upon in any interaction “are always already there in the always already-up-and-running communal linguistic practices into which I enter as a young one” (Brandom 2009: 73). Through the provision of these artefacts, every linguistic interaction is mediated by the concepts of the wider community.
If Hegel’s idea of Recognition is taken out of the context of his whole method it is easily misunderstood, and taken to be an unmediated binary relation between two individuals, but this is never the case; interactions between subjects are always mediated. As Hegel states at the very beginning of the Logic: “There is nothing, nothing in Heaven, or in Nature or in Mind or anywhere else which does not equally contain both immediacy and mediation” (Hegel 1816/1969: §92). Analytical philosophy, and all varieties of interactionism and recognition theories, systematically ignore this maxim of Hegel’s, which characterises his entire corpus. Mutual understanding even between strangers, apparently unmediated by common language or custom, is possible provided that each person can produce something which the other person needs. As participants in a shared culture there are concepts which are “always already-up-and-running.” This mediating element is something not created by the interaction (although every interaction maintains and modifies the culture). The mediating structure exists independently of any single interaction and is a ‘larger’ unit, being a property or aspect of the entire community of which the partners to interaction are a part. Concepts belong to this larger unit, and are evoked in the interactions and thinking of individuals as mediating elements. This stands in contradiction to Brandom’s efforts to found his inferentialism and his reading of Hegel exclusively in actions. It is as if actions and interactions (such as uttering a proposition, recognising another individual, committing oneself to a concept, etc.) can exist prior to and independently of the cultural constellations and social formations which mediate individuals’ actions and from which actions draw their meaning.
When Brandom ventures that “the unit of meaning should be taken to be a whole theory, not just a single sentence” he is admitting that a larger unit of analysis is required in order to make inferential actions intelligible. A “whole theory” cannot be conceptualised as a collection of propositions, any more than a human being can be conceptualised as a collection of molecules or a nation as simply an agglomeration of individuals. To grasp a “whole theory” one must understand the unifying principle which makes it a whole theory. This is absent from Brandom’s work. Such units are beyond the horizon of his theory. I will come presently to the rare occasions on which he ventures beyond this horizon, but generally speaking, the source of meaning lies outside his field of vision, and consequently one must conclude that inferentialism as Brandom has developed it must fail. There has to be some social fabric. Communities cannot rest solely on unmediated interactions between otherwise isolated individuals, and in his effort to prove otherwise, Brandom locates himself squarely in the analytical tradition, sharing perhaps its most characteristic blind spot.
Brandom’s commitment to holism goes only so far as the door to the academic neighbouring department. It is not so much Gottlob Frege’s Begriffsschrift which is the foundation of analytical science, but the agreement of everyone to pursue a career within their own disciplinary boundaries, without reference to what is happening in any other department. Philosophers must stay out of social theory and psychology, and vice versa. That is the essence of the analytical tradition. But what passes muster as a theory of logic cannot get past first base as social theory.
Concepts are products and vehicles of reasoning, that is true. But concepts can only exist as formations of human psyches. And as we have seen, concepts are essentially the cultural products of whole communities, and sustained by those communities. Concepts are not simply the function of individual human beings. It has to be granted that an understanding of concepts as norms and vehicles of inference is necessary to the understanding of what a concept is, since concepts are essentially both psychological and societal entities. But a theory of concepts which pays no attention to the social and psychological existence of concepts is untenable. But this is exactly what Brandom does. He laments, in the final section of “Reason in Philosophy,” that cognitive psychology has not investigated whether the development of concepts in children replicates his hierarchy of the complexity of concepts. Well, why not spend an afternoon with a child psychologist and ask? The development of concepts in children has been studied, and not only do children not replicate Brandom’s schema, but there are very highly developed theories of the development of concepts which demonstrate different genetic processes. If Brandom thinks there is reason to believe that the development of concept-use in children should mirror the structure of their semantic complexity, perhaps the way concept-use actually develops would give him cause to reconsider the significance he attaches to this hierarchy of conceptual complexity? Would it give him cause to reconsider his whole theory of concepts? He says that it is ignorance on the part of psychologists that they have not investigated the basis for his philosophy in psychology, but isn’t it more reasonable to look to ignorance on his own part that he has developed a schema for the development of concepts without taking the trouble to enquire as to how it actually happens?
The problem of the origin of concepts in social formations and Brandom’s lack of interest in investigating this origin is even worse than his lack of an active interest in ontogenetic development. In “Articulating Reasons” there is one line in the whole book touching on the societal origin of concepts. In the context of pointing out that virtually every sentence is unique, he says: “The linguistic community determines the correct use of some sentences, ....” How? He does not seem to realise what a problematic statement this is. There is a vast literature on the topic of the social origin of ideas, and the social conflicts tied up with the process of meaning-determination. But Brandom is either uninterested or unaware of these issues, systematically taking “society” to be a homogeneous and integral whole. And on the basis of his own failure to enquire into the origins of meaning, he simply takes concepts as given data. In “Reason in Philosophy” there is a chapter devoted to how “conceptual contents might be attached to states and performances.” He opts for what he calls a functional approach:
it is the activity of the system itself that establishes the conceptual contentfulness of the states it exhibits ... the system itself implicitly takes or treats its own states as contentful, and thereby makes them so. ... some vehicle ... for instance, sign-designs as marks or noises, performances as bodily movements, states as voltage distributions or neurophysiological conditions. ...
Leaving aside the efforts in the above passage to subsume human social life under a broader class of natural and mechanical processes, his claim is that “the system itself implicitly” determines the content of propositions. What system? How? Under any interpretation this claim is either empty or circular. We are concerned with how propositions acquire conceptual content, and Brandom has devoted his career to studying concepts but he stops just at the point when he would be obliged to enquire into just how conceptual content is created and vested in linguistic vehicles. This is untenable.
Brandom focuses his energies on how concepts regulate the reasoning process by acting as norms for judgment, but even within this domain of the investigation of norms he is acting blindly, because he considers only the end product of a long cultural and historical process of the development of norms. It is only possible to make sense of a complex whole (such as the system of norms operating within a community) by coming to understand it as the outcome of a long, conflictual process of development. To be more precise, norms must be understood both diachronically and synchronically, both as the outcome of a certain process of development (which shows why it is done this way and not that way), and structurally. The fact is that any real social formation is what it is as the result of certain historical experiences and social problems, and this is encoded in norms reflecting the metaphysical beliefs and ethical precepts that the social formation has adopted. For example, when a judge makes a determination of some dispute, they have recourse to precedents and it would be quite impossible to understand their determination, without having access to the precedents (as well as the social context in which the judge deliberates). Norms have a history and not only is it impossible to understand a norm independently of its history, it is actually impossible to understand norms in general, without studying the historical process which fashions norms. Brandom believes that norms of belief primarily regulate processes of inference and are properties of propositions. In other words, that norms regulate the business of doing philosophy. Norms and inference may be the subject matter of philosophy (I think Brandom is right on this), but philosophy is not primarily the subject matter of norms and inference. What is or may be the subject matter of norms, can be resolved only by a study of the process of the historical and cultural formation of norms and reason, outside of philosophy. Otherwise, to the philosopher, everything seems to be internal to philosophy. It is not just a question of “placing both within a larger historical developmental structure” (2009: 81), for the content of norms is in that “larger historical development,” beyond the domain of philosophy. The development of norms is the left to be the business of the History Department.
Brandom has however his own theory of history, including the history of norms. He thinks that norms are worked out by individuals making explicit what is implicit in the existing norms, as outlined in the narrative about judge-made law. Hegel’s concept of recognition is also appropriated:
So the process that synthesizes an apperceiving normative subject, one who can commit himself in judgment and action, become responsible cognitively and practically, is a social process of reciprocal recognition that at the same time synthesizes a normative recognitive community of those recognized by and who recognize that normative subject: a community bound together by reciprocal relations of authority over and responsibility to each other (2009: 70).
These two ‘models’ (judge-made law and recognition) are extremely rich, but it does not suffice in forming a theory of history to take a ‘model’ of interpersonal interaction involving one or two individuals, and then simply declare that historical processes involving human beings en masse are to be understood according to this ‘model’. History is history, and specific concepts are required to understand its processes, which may be different from the processes which are manifested in the interaction between just two people. Most particularly, when two people interact, they must perforce have recourse to already-existing language, concepts and other norms, developed, maintained, transmitted, concretised and objectified in societal processes involving human beings en masse. Norms arise from the cultural-historical process, which entails the modification and inheritance of material culture from generation to generation. Person-to-person interactions are subject to processes and norms different from cultural-historical processes, which remain, for Brandom, a closed book.
Overall, Brandom’s inferentialism is to be welcomed for placing concepts at the centre of philosophical enquiry where they belong, and rejecting the barren representationalism which has dominated analytical philosophy and its application in cognitive psychology. His emphasis on concepts as possible conclusions and premises for inference is surely correct, reflecting what it means to understand a concept. Although I have not dwelled on Brandom’s conception of norms, inasmuch as it unifies norms of belief, norms of desire and semantic norms, this too is a suggestion that deserves to be taken up.
The real problems with Brandom’s theory are two-fold. Firstly, in common with all analytical philosophy, he tries to address the problems of the human condition without venturing outside the narrow confines imposed by disciplinary boundaries, and substitutes for the important connections with other disciplines, uninformed guesses unworthy of science. Secondly, and most importantly, he has appropriated a version of Pragmatism which is blind to mediation. The appropriation of Hegel and Kant via a Pragmatic reading is surely the most fruitful approach to philosophical problems of our times, but pragmatics – the study of the practical aspects of human action – always entails the use of artefacts which are already-existing products of the wider society. It is these material artefacts (including words and symbols) which are the real bearers of the culture accumulated by a community down the generations, insofar as living human beings continue to use them in their activity. Omit these mediating elements and you are left with the atomism which is so characteristic of liberalism and analytical philosophy.