Wittgenstein and Social Practices

I was fascinated to discover Brandom’s 1976 dissertation, which overall is an original reconstruction of the key themes of the classic American pragmatists’ approach to the theory of knowledge. A number of Brandom’s own characteristic themes are already in evidence here; others have yet to be developed.

The first chapter elaborates the basis for Brandom’s later oft-repeated but rather telegraphic references to the great analytic philosopher Ludwig Wittgenstein (1889-1951) as a pragmatist, an identification that Wittgenstein does not make himself. Brandom argues that the main theme of Wittgenstein’s posthumous work Philosophical Investigations (1953) is a pragmatist account of knowledge, which aims to be a third way that is neither objectivist in the manner of Wittgenstein’s other main work Tractatus Logico-Philosophicus (German ed.1921; English tr. 1922), nor subjectivist in the manner of the Cartesian and phenomenalist traditions.

Brandom sees late Wittgenstein as offering a more incisive argument for the centrality of social practices in an adequate account of knowledge than any of the “official” (self-described) pragmatists Pierce, James, Dewey, and Mead. He aims to provide the clear account of what is meant by social practices that the canonical American pragmatists and even Wittgenstein did not. Brandom himself here will still rely on a broad notion of community to ground the justification of claims, which is less sophisticated and less adequate than his later account based on Hegelian mutual recognition.

“But what is a social practice? I think that most of the misunderstanding and undervaluation of the pragmatists (Wittgenstein included) stems from their failure to give a clear and unambiguous answer to this question” (pp. 8-9).

“Social practices are best understood in terms of a criterial classification of things…. There are three basic criterial categories. First of all, there are things whose criteria involve only the attitudes and behavior of an individual person. Sensations are things of this kind…. Following Rorty, I call things for which we accord this sort of criterial authority mental. Second, there are things whose criteria are the attitudes and behavior of groups and communities of people. A particular motion is a greeting gesture for a tribe just in case they take it to be one…. I will call this kind of thing social practices…. Finally, there are things whose criteria of identity are independent of the attitudes or behavior of any individual or group…. I call this kind of thing objective” (pp. 10-11).

The second chapter of the dissertation builds on Rorty’s distinctive and original account of the role of “incorrigibility” in Descartes. I’ll address this in an upcoming post.

In later works, Brandom has often rather summarily dismissed definition and classification, as made obsolete by Hegelian recollective genealogy. I think his dismissals go too far, because they suggest that definition and classification are only ever applied in an objectivist manner, as if they simply fell from the sky. Such a suggestion does not adequately recognize the profound dialecticization that identity, definition, and classification already undergo in the hands of Socrates, Plato, and Aristotle. For them, definition and classification are anything but taken for granted. It is precisely open inquiry into criteria of identity, definition, and classification that they commend to us.

“The criterial division is simply into things that are whatever some one person takes them to be, things that are whatever some community takes them to be, and things which are what they are no matter what individuals or groups take them to be…. Put another way, meanings, the things that we grasp when we understand something, are taken to consist of social practices by the pragmatist, of mental particulars by the subjectivist, and objective facts by the Tractarian [early Wittgensteinian] objectivist…. In the rest of this chapter we will examine perhaps the clearest sustained argument for the pragmatic rendering of meaning and understanding in terms of social practices, Wittgenstein’s Philosophical Investigations” (pp. 11-12).

In later works, Brandom has usually discussed “taking” things to be thus-and-such in Kantian terms. Here it takes on a pragmatist coloring, and Kant is not mentioned. But the founder of pragmatism, Charles Pierce (1839-1914), was deeply influenced by Kant (and expressed an affinity for Hegel as well).

“The pragmatist must be able to explain how, by engaging in various social practices (which are things of the second kind, over which the community has complete dominion), we can come to express, make claims, and have views about objective matters of fact (which are things of the third kind, independent of the attitudes of any community)…. Pragmatism as a view of human functioning stands or falls with the project of giving some such account. No pragmatist, including Wittgenstein, has explained what it is about our linguistic social practices in virtue of which they are appropriately taken to involve claims about objective things” (pp. 12-13).

This concern with the constitution of objectivity — indeed the treatment of objectivity as something constituted rather than as something given, never-you-mind how, is a very Kantian sort of problem. Kant does not really address the social aspect that is in the foreground here though.

Next he gives an overview of the argument he will be attributing to Wittgenstein.

“There are three basic lines of argument running through the Investigations, corresponding to three ways which one might think of to eliminate the reference to social practices in talking about meaning and understanding in favor of things of the other two kinds, objective and mental…. In the argument we consider first, Wittgenstein examines the Tractarian notion that meanings are objective things, which objectively determine the correct applications of expressions. The second argument we will consider examines the Cartesian notion that meanings are mental things (such as images), which objectively determine the correct usage of expressions…. Third, we will consider the so-called ‘private language argument’, which I take to be an examination of the view that meanings, whether mental or objective things, determine correct occasions of use of expressions by a mental process. The argument in each case will try and establish the same claim, namely that whatever sort of thing one imagines as intervening between an expression and its use or application in concrete circumstances, that use or application must be taken to be a social practice” (p. 14).

“What kind of thing Xs are (meanings, uses, understandings in this sense) is a matter of the criteria which determine whether something is an X or not…. Wittgenstein will develop an answer to the question of whether something counts as meaning or understanding something (or learning, remembering, thinking, reading it) by creating a series of analogies (‘family resemblances’) to other familiar activities which share the criterial properties of social practices” (p. 16).

“The objectivist takes meanings to be, not uses or social practices, but objective things like words…. [S]ocial practices admit of a sort of indefiniteness or vagueness which objective things do not. Thus Wittgenstein begins his attack on the view that meanings are objective things and determine the application of expressions objectively (‘according to definite rules’) by asking whether the use of a sentence or a word must be everywhere determined by rules in order for the expression to have a meaning, or for someone to understand it” (p. 17).

In passing, I would point out that Plato and Aristotle too treat rules with a healthy skepticism. That is why Plato insists that good government requires philosophers, and Aristotle builds his ethics on deliberation and practical judgment.

“For some performance to count as an instance of a social practice is for it to be accepted as such by the relevant community. And this means that there can be a social practice without its being the case that for every imaginable performance the community has decided in advance whether it would be acceptable or not. There is a social practice as long as there is sufficient agreement about the cases which actually come up…. In just this respect social practices differ from things of the third kind, which are independent of the attitudes of particular communities” (p. 18).

The very notion of “practice” has an inherent open-endedness.

In Brandom’s later terminology, the constitution of normativity has an inherent dependency on attitudes. While it acquires a kind of (always qualified) objectivity, it does not originate as something objective.

“There is no vagueness about whether, for instance, a given word appears in the rule or not. Insofar as this sort of thing is left vague, one has not specified a rule or expression at all. The question is whether for an expression to have a meaning (or be understood) its application has to be similarly objective and definite, whether the syntactic objectivity must be matched by semantic objectivity. Wittgenstein attacks this sort of objectivism by pointing to the vagueness we tolerate in the application of expressions, arguing that the use of an objective rule or expression is a social practice, that is, that the criterion of successful application is its actual functioning in the community” (p. 19).

“The attempt to eliminate social practices generates a regress, for no rule generates its correct application to concrete circumstances by an objective process” (p. 23).

This point about rules is fundamental. While we can always try to express things as objectively as possible by formulating rules for the application of rules (and more rules for the application of those rules, and so on), this is at best an infinite regress, and there is always a remainder.

“An object, such as a rule, can determine a practice only if there are other practices, e.g., of responding to the object, in the community…. Wittgenstein explicitly draws the lesson that social practices, as things of a different kind from objective things, are ineliminable in accounts of this sort” (p. 24).

In a nutshell, this is what justifies Brandom’s characterization of Wittgenstein as a pragmatist.

“The social practices which are being contrasted with objective things in these passages are not strange or spooky things, and they are certainly not subjective” (p. 25).

At a popular level and even in some philosophical discourse, “pragmatism” is often treated as a kind of subjectivism. The full basis for rejecting this has not yet been elaborated, but it will involve a contrast with Rorty’s notion of incorrigibility.

“Mastering the practice is not a matter of following any set of rules, but rather of behaving in a way acceptable to the rest of the community. Rules may play a role in this, but need not. This line of thought can be brought to bear against the notion that cognitive functioning consists of manipulation of things of the first kind, mental events or processes” (p. 26).

“Mental things do not have an essentially different relation to such applications than linguistic rules did…. [U]nderstanding the meaning of an expression does not consist in some sort of mental grasp” (p. 27).

Again, part of his argument depends on Rorty’s innovative re-characterization of what it is to be “mental”, about which we will see more in a future post.

“The discussion of objective rules goes over to mental events and processes quite unchanged…. The connection between being in a certain mental state and understanding an expression is empirical. The criterion for being in the mental state is, roughly, that one sincerely think that one is, while the criterion for understanding an expression is that one be able to apply it in ways that the community accepts as correct…. The understanding one has of the meaning of an expression is so far from identical to a mental state that the state only becomes sufficient evidence for the understanding in virtue of a social practice of taking it to be so…. The mental functions here only as the invisible inward sign of a visible outward (i.e., social) grace” (pp. 28-29).

“Thus it is clear that the meaning of an expression may not be taken to be a kind of mental state which is elicited by the expression in the members of some population when they understand the expression, and which then objectively determines the use they make of that expression. There must be a social practice of applying the expression…. Wittgenstein is not claiming that mental states have no role to play in this process. He is claiming that they cannot replace the social practices of applying linguistic expressions…. [W]hat makes the performance correct is its consonance with the practices of the rest of the community, and this cannot be a matter of mental or objective processes” (p.30).

“The famous private language argument seeks to show that by the very act of making the language mine own, I must make it a poor thing…. The argument Wittgenstein makes is that ‘ “obeying a rule” is a practice. And to think one is obeying a rule is not to obey a rule. Hence it is not possible to obey a rule “privately”: otherwise thinking one was obeying a rule would be the same thing as obeying it’ ” (p. 32).

This is a decisive point: a “private rule” — one in which the one acting is identical with the one authorized to evaluate that one’s compliance with the rule — is no constraint at all. It is indistinguishable from sheer arbitrariness.

“For one’s authority over the character of his mental states and processes is complete. One is incorrigible about such matters, that is what it is for them to be mental according to our stipulated usage. What is wrong with such mental rules? What is wrong is that they cannot, in principle, be transgressed. Accordingly, they do not establish any boundaries between correct and incorrect usage, not even the vague boundaries induced by social practices” (p. 33).

In a footnote, he says “I am not taking into account the strand in [Wittgenstein’s] thought which would deny any cognitive status to incorrigible first-person avowals” (p. 33n). Apparently Brandom does not (or at least, at this time did not) consider this to be part of Wittgenstein’s “better wisdom” that he wants to emphasize.

This matter is a bit delicate. I suspect Brandom wanted to save the appearances by not throwing out a common-sense acceptance of first-person insights, but I am inclined to think that this reported denial by Wittgenstein is a necessary consequence of the argument. Whatever is in principle immune to criticism ought to have no standing in serious discourse. As Habermas says about the entry conditions for ideal speech situations, each participant must willingly submit to questioning and criticism. The delicate part is that the concern to save the appearances of common sense is also a legitimate one. But I think the legitimate concern to “save” a common-sense validity for first-person avowals and reports does not require giving common sense a strictly cognitive status.

This comes back to the Platonic distinction between knowledge and opinion. At odds with the mainstream tradition that “knowledge = justified true belief”, I maintain that there are many things that are legitimately considered to be objects of well-founded belief, but that still do not strictly qualify as knowledge in a strong sense.

He quotes Wittgenstein, “One would like to say: whatever is going to seem right to me is right. And this only means that here we can’t talk about ‘right’ ” (p. 34). This is a passage Brandom frequently cites in his later works. I think it similarly means that here we can’t talk about “knowledge” in the strong Platonic sense that I uphold.

Brandom adds more support for my argument in the following.

“In a social language, the community which determines whether a given utterance is a correct use of an expression is different from the individual who utters the expression. There is accordingly room for a judgment of incorrectness. But in the case we are imagining, the individual who produces the utterance and the one who judges its correctness with respect to the original rule or definition are identical. There can be no check of whether a given performance is in accord with the rule which is independent of the performance itself. Indeed, there can be no evaluation which is not identical to the performance” (ibid).

“One may wish to call an activity with no rules whatsoever a game, but one may not then go on to claim that there is a difference between playing it and not playing it” (p. 35).

Next in this series: Seeming, Trying

Love’s Intellect

The main theme here is an unexpectedly close relation between love and a broadly Aristotelian notion of intellect. We will also see another perspective on the crucial Aristotelian concepts of potentiality and act, and perhaps shed further light on Aristotle’s telegraphic remarks about how we have knowledge of “privations” or negations of things.

I’ll be devoting two posts to the small pamphlet Intellect d’Amour (2018), introduced by the great contemporary scholar of medieval philosophy Alain de Libera, in which leading specialist in Averroes and Latin Averroism Jean-Baptiste Brenet translates an Italian essay by Giorgio Agamben, and presents a related essay of his own. This post deals with Agamben’s part.

While I have little sympathy for Agamben’s quasi-Heideggerian reduction elsewhere of “Western metaphysics” as a whole to a fundamental error for which Aristotle is supposedly to blame, his essay here is focused and interesting. The medieval Italian poet and philosopher Guido Cavalcanti (1255?-1300) — a close associate of and influence on the great Dante Alighieri, author of The Divine Comedy — is now generally understood to have been inspired by Italian Averroist philosophy (see Italian Aristotelianism). The dominant interpretation of Cavalcanti, however, has been that of Bruno Nardi, who emphasizes a fundamental discontinuity between Cavalcanti’s exaltation of love in both its spiritual and its erotic dimensions, and his Averroist views on intellect. Agamben, Brenet, and de Libera all seem to agree in turning this interpretation on its head.

Dante and Cavalcanti are both commonly associated with the historically shadowy group of poets known as the fedeli d’amore (love’s faithful). The fedeli were apparently influenced by the poetry and music of the troubadours, who developed the Western medieval tradition of courtly love, and also used it as a spiritual metaphor, somewhat along the lines of Plato’s Symposium. Sources of the troubadour tradition are disputed by scholars and likely were multiple, but an Arabic or specifically Sufi element has been repeatedly suggested.

Agamben’s essay points out that a particularly mysterious phrase bianco in tale obiettò cade (something like “white falls into this objection”) in Cavalcanti’s poem Donna me prega appears to be intended to recall the Latin cecidit albedo in exemplari in Averroes’ Long Commentary on Aristotle’s On the Soul. In the poem, Cavalcanti has just said that love cannot be understood in terms of vision.

Agamben notes that in the passage where the corresponding phrase appears in Averroes, Averroes is more broadly addressing how we know the privations of positive terms. In the context of vision, Averroes poses “blackness” as a privation of “whiteness”, somewhat like modern science calls blackness an absence of color. Agamben quotes de Libera’s remark in his partial French translation of Averroes’ work that “Every privation is effectively known negatively, by preliminarily positing something and then negating it” (p. 13, my translation throughout).

The “white” in Cavalcanti’s poem recalls the white by which the black is known as a privation. Following the passage Cavalcanti refers to, the Latin translation of Averroes says “And it is necessary that this faculty of knowledge perceives the privation in perceiving itself as being in potentiality, when it is in potentiality, since it perceives the one and the other of itself, that is to say being in potentiality and being in act. And such is the case with the material intellect” (quoted, p. 19).

Here we have among other things a fascinating connection between self-apprehension and negation. I imagine Hegel nodding in approval here.

“Averroes affirms that, in knowing obscurity, the material intellect knows itself insofar as it is in potentiality, and that, reciprocally, for the material intellect, to understand itself is equivalent to knowing obscurity and privation. One sole and same faculty — the material or possible intellect — knows obscurity and light (obscuritatem et lumen), power and act, form and its privation. As de Libera notes, this signifies that ‘privation is attached to the very essence of intellect’, and indeed also — the consequence is inevitable — that obscurity — non-thinking in act — is an integral part of intellect and is consubstantial with it ” (pp. 19-20).

Here I think also of Socrates’ comment that wisdom involves recognizing what we do not know. In a sense, this kind of recognition of privation is just as much a part of knowledge as any positive content.

“The idea that obscurity, which for him constitutes an essential part of the amorous experience, in no way implies an exclusion of knowledge, could not but fascinate Cavalcanti. On the contrary, since one sole and same faculty — the possible or material intellect, where love has its dwelling place — knows equally well the darkness and the light, equally well the form and its privation, having the amorous experience of obscurity thus also necessarily signifies having the experience of the pure power of intellect” (p. 20).

In passing, Agamben refers here to his major thesis in this essay, that the material intellect is where love has its abode for Cavalcanti. We also see an example of the Aristotelian thesis that rational knowledge of one of a pair of opposites necessarily entails knowledge of the other.

“In this perspective, it is possible to divide the philosophers into two great classes or families. According to the one group, what defines humans is the act of thinking, and the latter are indeed like the angels, always in the act of thinking; according to the others (the Averroist tradition in which Cavalcanti as well as Dante are inscribed by right), what defines humans is not the act, but the power of thinking (humans do not think continually, which is to say they think in an intermittent way — not sine interpolatione [without interpolation], as Dante says” (pp. 20-21).

In our thinking, we who are not pure act depend on this “power of thinking” that begins as something external to us.

Agamben recalls that Averroes calls the material intellect a fourth kind of being (neither form, nor matter, nor a composite). “It is this perception of its own obscurity by the material intellect, essentially divided in its being, which becomes for Cavalcanti the place of the experience of love” (p. 22). “If the attempt to think, apropos of the material intellect, the existence of a pure power as a fourth genus of being leads Averroism to an aporia from which it is not easy to escape, it nonetheless furnishes, and precisely through that, the elements of another conception of subject than that which has prevailed since Descartes…. Otherwise said, Averroism thinks the subject as the subject of a power, and not only of an act…. Averroes suggests that the material intellect should be considered more as a place… than as a matter” (p. 23).

Agamben’s language seems overly loose here, in that it blurs together act in an Aristotelian sense and agency in a modern sense, but he nonetheless makes an important point. The “Cartesian subject” is indeed commonly conceived more or less exclusively in terms of its agency in the modern sense. Both the mechanist and the voluntarist dimensions in Cartesian thought mitigate against taking the key Aristotelian concept of potentiality seriously (and conversely, taking potentiality seriously makes both mechanism and voluntarism untenable).

The modern notion of agency is distinguished by the fact that from the outset, it is conceived as not having any inherent relation to a contrasting term like potentiality, that would condition and limit it. By contrast, Aristotle only arrives at the thesis that there can be such a thing as pure act after a long dialectical development, starting from the cases where act is only analytically distinguishable from potentiality. The common modern approach short-circuits all of this. Aristotelian act and agency in the modern sense are thus two different things. But Agamben correctly points out that any finite “subject” that is the subject of a conditioning power (or capability, as Paul Ricoeur more felicitously puts it) will be fundamentally different from the modern stereotype of a one-sided subject-agent.

“The great invention of Cavalcanti, of Dante and the other poets of love, is to situate love in the possible intellect without reservation. As the song says, and for once clearly, the possibile intelletto is the subject — or the quasi-subject (come in subietto) — the loco [place] and the dimoranza [residence] (the mansio [house]) of the veduta forma [form seen] that produces love…. [L]ove (as the ‘form seen’, veduta forma, which gives birth to it) is in the possible intellect as in its proper place” (p. 24).

The thought here seems to be that the feeling of love follows from the apprehension of a form or image as having characteristics that make it in itself lovable. According to Agamben, for Cavalcanti and Dante, the possible aka potential aka material intellect is the “quasi-subject” of love. At the same time, the possible intellect is not so much a discrete entity as a “place” that is not really a discrete place either, but a kind of ubiquitous structural relation.

Agamben cites a line from Dante speaking of “women who have love’s intellect”. He comments, “[B]etween intellect and love the connection is essential” (ibid).

“The modern specialists, each convinced that they think with their own brain — when to all evidence they on the contrary think according to the common paradigms imposed by the doctrinal system in which they are inscribed — experience such distress before the Averroist theory of the unique intellect, that they do not understand what should nonetheless be evident, namely that the speculative node of the question, so to speak the experimentum crucis [cross test] of Averroism, does not consist in the division between individuals and intellect, but — once the division is affirmed — in their conjunction, which the Latin translators render by the technical terms copulatio and continuatio. In this conjunction between the unique material intellect and singular individuals, the essential mediating function is accomplished by the intentiones ymaginatae, that is to say the phantasms of the imagination” (p. 25).

“Nonetheless, not only does the imagination operate as an intermediary between individuals and the intellect, but it is also for Calvalcanti the object and at the same time the subject of amorous passion” (p. 26). “Cavalcanti and the poets of love take the coincidence between love and imagination so far that they personify the phantasms” (p. 27).

This is saying that imagination is both the object and the subject of love, as well as what individualizes intellect.

Agamben points out that Averroes strongly underlines Aristotle’s statement that only the combination of intellect and desire moves us. “It is this singular fusion between intellect, desire, and imagination that it is also necessary to understand in the ‘love’s intellect’ of Dante…. Love is not a substance, it is — as [Cavalcanti’s] song says in its overture — an ‘accident’ that indetermines the three substances intellect, imagination, and desire, and puts them fiercely in tension” (p. 29).

“[F]or intellection to be ‘acquired’ and become ‘proper’ to the individual, in effect it does not suffice that the forms be imagined, but it is necessary that they be desired and willed…. [T]he phantasm by desire makes the intelligible proper to the subject. Thought belongs to me because it has been imagined and desired” (p. 31).

“The great invention of Cavalcanti and the poets of love is to make love the place par excellence of the adeptio [acquisition] of thought by the individual” (p. 32).

Whom and what we love are of decisive importance to who we are, as well as to what we hold to be true.

Agamben notes that Dante also drew political consequences from his Averroist view of intellect. He quotes Dante’s treatise on monarchy: “Since the power of thought cannot be integrally and simultaneously actualized by a single human or by a single particular community […], it is necessary that there be in the human genus a multitude through which the whole power is in act” (p. 33, ellipses in original).

Dante’s universalist aspirations distinguish him from Cavalcanti, who was deeply involved in factional intrigue. For Dante, “‘our’… accompanies and precedes ‘me'” (p. 34).