Back to Habermas Studies page Reason: reflexivity and rationalization
gary e. davis
October 2003 / March 2014 / March 2017

[This is a posting from the Spoon Collective which I wanted to liberate from that e-mail platform. I’ll revise it eventually, but it’s good as it is, I think. It was thoughtfully written as an essay draft, not improvised; but it was uneditable. I’ve done some editing and inserted some comment.]

What is reason?

Oddly, ‘reason’ doesn’t appear in the Index of On the Pragmatics of Communication (MIT Press, 2000), but there’s lots of references for ‘rationality’. Herbert Schnädelbach wants to distinguish the two, as indicated by the long quote at the beginning of “Some Further Clarifications of the Concept of Communicative Rationaltiy” (1996) with which Habermas takes issue.

The Stanford Encyclopedia of Philosophy doesn’t have an entry yet on reason, but it has one on moral reasoning, and there reason is indicated as an inferential capacity. This accords with the normal definition of reason:

Webster’s Unabridged Dictionary indicates a difference between reason and rationality, in its definition of reason “2 a (1) : the power of comprehending, inferring, or thinking especially in orderly, sensible, rational ways....

The power is orderly, sensible, and also rational, but not reducible to its rationality (capacity for accountability). Webster’s definition continues:

...: the ability to trace out the implications of a combination of facts or suppositions ... (2) : proper exercise of the intellective faculty in accordance with right judgment...

Reason is an ability, as well as a faculty that is exercised rationally. But there’s a difference between the ability / faculty and its rational exercise.

...a distinct cognitive faculty coordinate with perception and understanding : human intelligence or intellect (2) : the sum of the intellectual powers ....

This normal sense of reason accords with Dewey’s sense of intelligent action, employed by Putnam in his critique of Habermas. This normal sense of reason accords with the sense of reason in the vast field of cognitive science. And this normal sense of reason is also foremost in Schnädelbach’s concern, quoted by Habermas:

S: Other rational capacities include the capacity for testing reality (Freud), for learning from mistakes and errors (Popper), for solving problems in feedback-controlled action contexts (Gehlen), for purposively selecting means (Weber)—many other prominent examples could be added to the list; those I have mentioned simply cannot be accomodated in a schema of “justification” or of “the discursive vindication of validity claims” (Habermas) [OPC 307]

JH: What he sets up against discursive rationality is...the “reflexive character” of...expressions. For, of course, what we know, do, and say is rational only if we are implicitly aware of why our beliefs are true, our actions right, and out linguistic utterances valid.... (307-8)

It’s clear that there is a difference between (1) use of reason in the senses indicated by Webster, Dewey, and Schnädelbach; and (2) rationalization of one’s reasoning to and for others (or for one’s own self-assurance), i.e., accountability. Schnädelbach is making this kind of distinction while using ‘rationality’ for both kinds of sense, while JH is using his favored sense (2) of ‘rationality’ to characterize Schnädelbach’s point.

Yet, the rationality of X, in JH’s sense, is only as good as the reasoning to X (valuative, epistemic, or self-reflective determination) that backs a claim that is justifiable. Schnädelbach intends to capture this kind of point, it seems to me, in his notion of reflexivity:

S: [T]he trope of ‘reflexivity’ as the fundamental characteristic of rationality in general, can thus be rendered more precise with the help of the self-referential thematization of...performances in the perspective of the first person singular or plural; only she who is capable of saying ‘I’ or ‘we’ and of thematizing what she is or does, and of attributing it to herself, is rational. (308)

It may be that Schnädelbach’s sense of reflexivity unfortunately [JH} “places himself within the tradition of the philosophy of consciousness” (308); I know nothing more about Schnädelbach than what JH selects for quotation. But it’s not the case that reason must resort to a psychologism in order to not be assimilated to responsibility and accountability; there’s good reason to believe that JH is missing Schnädelbach’s point. In any case, it’s important to appreciate the difference between reason and rationality, especially inasmuch as one is interested in reflexivity. The field of cognitive science is not some vast proxy for a philosophy of consciousness.

Given that the nature of reason is claimed to be reflexivity, in some sense of this—not presuming the meaning of reflexivity, but stipulating that inquiring into what reason is is potentially fruitful and naming that “is” reflexivity—the question is: What is reflexivity as such? It’s a dimension of intelligence, whose hallmark might be called learnability. One’s intelligence essentially is learnability (analyzable into various components), which is the condition for the possibility of mental performance—Dewey’s “intelligent behavior” (hallmarked by creativity, Mead argues); and learnability can be a keynote of reflection. Reflection is not only an evaluative capacity premised on constant values; it, too, is a mode of learning. Reflexivity grows or advances or progresses in its metacomponential capacity for learning, which then contributes to the constitution of mental encoding, mapping, comparison, and inference that is normally associated with reason. (This view is based on the theory of intelligence developed empirically by Robert J. Sternberg, who has been the leading U.S. researcher into intelligence for a couple of decades.)

So, given the importance that I’m able to find in the notion of reflexivity, I’m glad that JH uses Schnädelbach’s point as an entrance into consideration of reflexivity. To understand reflexivity is to understand the nature of reason, inasmuch as growth and progress of reasoning—improved capacity for insight, problem-solving, and mental efficacy—hinges on reflexive learning.

JH: Since the linguistic turn,...we have good reasons for following a suggestion of G. H. Mead and explaining the self-relation of the knowing, acting, and speaking subject—that is, the relation of the first person to “herself”—on the basis of the adoption of the perspective of a second person “on me.” (308)

On the contrary, there are good reasons to not orient our sense of self-relation on such a reading of Mead. But this is a long story. In brief, healthy individuation (which begins in early childhood, of course) requires accomodative and attributional activities toward the world and oneself, as well as assimilative activities. JH focuses (in his reading of Mead, in TCA and PMT) on assimilative activities; he understands self in terms of assimilations of social images.

But early identity formation happens through a separation-individualization process wherein accomodation and attribution tend to prevail over assimilation; this is what makes separation processes individuatory. Growth of the capacity for self-monitored learning, problem-finding and solution, imaginative idealization, etc., depend on the strength of the intelligent accomodation of perception (across all modes of experience) and language, in which socialization is secondary. Initiative is more important to individuation than socialization (JH tends to have a passive view of individuation). The basis of individuation is not in socialization as such.

Unfortunately, many people are well-socialized while being not highly individuated. Independent mindedness is good for democracy and for creative lives, but much of modern society is conformist (not to mention premodern formations throughout modern society). Independent mindedness, by the way, is not as such ego-centric or narcissistic. Habermas tends to have no discursive sense of independent mindedness (or high individuation) apart from his critique of subjectivism—though, ironically, he exemplifies a highly individuated mind (captured in the designation of some view of his as especially “Habermasian”).

Above, JH represents “me” as a second person perspective. Elsewhere, JH recognizes the sense of “me” that is indicated by William James and which has been confirmed by empirical research, which is very different from JH’s “me” here. (I cited this research in my discussion, late-2001, of JH’s case against liberal eugenics; the research was reported in Self-understanding in Childhood and Adolescence, William Damon and David Hart, Cambridge UP 1988).

“I” is distinct from “me” analogously as performing an action is distinct from my recognizing that there is my performing (“me” isn’t primarily derived from recognition by an other; this is dramatized by an infant’s amazed discovery that the hand in her face is controllable, which is a developmental milestone in early infancy, discovered more quickly by some infant temperaments than by others. Daniel Stern has shown how self “concepts” work in early infancy: The Interpersonal World of the Infant, Basic Books 2000, where rigorous clinical research shows how self concepts precede representations by the [m]other).

What we ordinarily understand as self-identity is what James (and persons generally, verified by Damon & Hart) means by “me”. (I use ‘concepts’ here in the sense of basic abilities, in accord with Ruth Garrett Millikan’s brilliant view in Clear and Confused Ideas, Cambridge UP 2000.)

“I” does. “I” shows in pre-represented doing; it’s the in-the-flow ongoingness, while “me” is the cognition of that. This ongoingness can be structurally theorized (which is what Stern does for infancy and Damon & Hart do for childhood and adolescence).

“I”’s implicity is the expressed self concept of one’s lifeworld background, wherein action-specific implicity [March 19: enacting the act] is immanently displaced by self-representation of it (JH inaccurately characterizes lifeworld features as being “annuled” by representation). [March 2014: There’s a 3-part differentiation here: [1] background Self (which has no representation; it shows as phenomenality of experience, the phenomenological “object”/other), [2] expressing self in enaction (lived, pre-denotational “I,” be-ing the enaction), and [3] representation of it or “me” with use of ‘I’ (the normal denotation of “myself,” the agent component of a represented action). This is easily confusing because representation of enacting is ordinarily one remove from what is represented. An overt notion of Self is a conceptual construction, e.g., a phenomenological conception, traditionally as transcendental subjectivity; or Heidegger’s Da-sein; or conceptions of the self-efficacy of an ontogenic capability. I’m continuing here, 2003, with an improvised version of this 3-fold selfhood.] [March 27, 2017: A more refined sense of this requires explication, which I will provide and link from this point.]

To make a very long story short, identity in postconventional understanding includes integration of sociopersonal relations—the integrity of one’s personhood—but is basically more than this: This integrity is nested in the temporality of self-understanding (background self concept or selfhood), which includes self-reflective “me” meaning—including first person expressions (though “I am” of self-knowledge, i.e., the me, is so much more than what’s expressed)— which has all along (through development) contributed to interpersonal relations and has accomodated those relations to individuation.

The more that this is the case—i.e., that self-actualization guides individuation—the less that individuation is relativized to interpersonal relations. This is especially noticeable with talented persons; they seem obviously very individualized.

The individuality of the highly talented person is an extreme of the individuation process itself, so it’s appropriate that individuation be theorized relative to (in the interest of) highly talented persons, which is typified by research into creativity (cf. Howard Gardner, Intelligence Reframed, Basic Books 1999). Highly talented individuation is ideal for understanding the nature of reflexivity, yet such idealization can rely on research on how such individuation really goes (i.e., the idealization for given cases can be based in a realism about highly talented individuation).

Reflexive learning is best understood relative to talented self-reflexive learning, which involves a guiding sense (for the learner) of self-reflexivity in general (relative to one’s lifeworld or life path of interests, values and engagements).

So, I want to look at Habermas’ sense of self-reflexivity relative to the ideal-typical sense of self-reflexivity that I’ve introduced (but barely explicated) above. I want to further develop an approach to self-reflexivity relative to Habermas’ discussion (though fundamentally contrary to it—but, again, for the sake of furthering my intent to complement his sociocentrism, not discard it).

Later, I will show, in terms of recent research (clinical and empirical) how a sociocentric reading of Mead is contrary to Mead’s sense of individuation (growth toward independent mindedness and creativity), as well as furthering the above case that JH’s sense of individuation is inaccurate and inappropriate. [March 2014: As best I can recall, I didn’t follow through—probably because list subscribers didn’t express interest.]

Continuing on—directly after the JH passage quoted 15 paragraphs above—it’s not the case, then, that:

JH: Correspondingly, the reflected self-relation distinguished by Schnädelbach as the fundamental characteristic of rationality would be dependent on the relationship between participants in argumentation: ... (308)

JH is simply assimilating reflexivity to justification. Reflexivity is fundamentally a discovery activity, a way to gain insight—what JH calls “world disclosure” later in “Some Further Clarifications.” JH’s rationalism depends on a linguistic relativism that is appropriate to social theory, but doesn’t pertain to reason altogether. A mind is enabled through linguistic language—more than through other modes of intelligence—for the sake of employing linguistic language (being intelligent linguistically), but language itself has no intentionality; minds employ, for the sake of— whatever, in “terms of” whatever purpose one gives oneself, in light of whatever motivating insight or desire, etc.

...there would be no reflection that could not be reconstructed as an inner discourse. (ibid.)

As,” surely; but it would be reductive to reconstruct reflection as an interpersonal discourse (which he does, later). ‘Discourse’, in JH’s statement above, should be (to be fair to him) understood in his technical sense of this, in which case reflection becomes basically a matter of overt hypothesization, explicit counterposing of stances, delimitations of all relevant possibilities, etc. But that’s modeling reflexivity as such on methodic social inquiry, which is inaccurate (I’ll show).

One can learn to engage in discursive reflection, in JH’s sense of this (technical deliberations require this); but relexivity belongs to individuation altogether (across interests, thus across modes of specialized discourse), just as does cognition, from stages of early child development onward.

In adolescence, reflexivity is especially social due to the character of identity formation in this era of development. But the relatively sudden ontogenic capacity for algebraic conceptualization and other abstract abilities is no matter of social internalization. Emerging from mid-adolescence, the talented individual emerges from sociocentric identity formation in a venture of field mastery (and identity formative reflexivity relative to this) that is dramatically distinct (in terms of the individualization of intensive focus) from the social domain (Howard Gardner, in sync with R.J. Sternberg, has elaborated this approach—talent vs. field vs. domain—in several influential books, which is summarized in Intelligence Reframed, Basic 1999). In terms of JH’s model, the interplay of personality and culture in talented individuation prevails over the interplay of personality and society; JH mistakenly tends to either give prevalence to society over culture or assimilates culture to society.

Earlier in JH’s career, he always characterized reflection as an internalization of “dialogue roles,” which has more plausibility across developmental eras than assimilating reflexivity to inner discourse. Merging both notions, reflexivity is for JH generally an internalization of dialogue roles that idealizes inner discourse. But this more plausible view—from dialogue to discourse—still doesn’t accord with the fact that accomodation, learnability, and constructibility prevail over assimilation, communicability, and accountability in reflexivity that basically counts, i.e., reflexivity that discovers, gains insight, and solves a difficult problem (which is not to exclude the importance of social problem-solving for the material “processed” by the one, usually, who makes the discovery, gains the insight, or solves the problem).

At the philosophical end of the spectrum (of developmentality), thinking is discursive, but not (I would argue) like dialogal argumentation. The “inner discourse” that gains insight is like—-well, a “poetic” [March 2014: i.e., conceptually specialized] use of language allows for more appropriate rigor about this than does JH’s figuration of an argumentation between discrete stances. This might seem obvious from trying to understand the difficulty of an original mind. Do you argue with a difficult text in coming to understand it? Is the path to cohering thought contentious? No. Critique is contentious, but coming to understand the other (text) is fundamentally disclosive. So, too, for fundamental reflexivity.

(By the way, a possibly inherent textuality of reflexivity is going on for Derrida. JH’s difficulty with Derrida might relate importantly to JH’s conception of reflexivity. This is the impression I get when I read “On the Distinction between Poetic and Communicative Uses of Language, ” OPC, ch. 9.)

JH: I do not consider the proposal to reduce rationality to a disposition of rational persons [to be] promising.

Who’s proposal? JH’s own quotation of Schnädelbach (307) counters the claim that such a proposal is Schnädelbach’s, so JH is missing the point. [March 2014: A dyadic sense of reason as including a sociocentric mode of rationalization, used in interpersonal justification and shared inquiry is not a reduction of “rationality” to subject-centered reason. Schnädelbach was overtly posing a dyadic view.] The point is that Schnädelbach is emphasizing a primacy of “reflexivity” for “rationality” that can be characterized in terms of the richer normal sense of reason which includes JH’s sense of rationality (accountability), but which is basically much more than that.

JH knows that he’s missing something:

However, this [JH’s critical commentary] does not invalidate Schnädelbach’s objection to my privileging of the discursive rationality emboded in argumentative practices. I will accept Schnädelbach’s point of criticism and, in the following, assume that we use the predicate ‘rational’ in the first instance to refer to beliefs, actions, and linguistic utterances, because.... (308)

Wait a minute. Schnädelbach’s concern wasn’t primarily about the use of a predicate, rather about what “rationality” more fully is, basically relative to “reflexivity”. Yes, JH misuses the predicate because rationality is generally more than JH admits. But the issue, I think, is that we use ‘rational’ in the sense of reason, which basically involves reflexivity. It’s not the case that “we use the predicate ‘rational’ in the first instance to refer to beliefs,” etc. We use the predicate primarily to refer to capacities for belief formation, capability for fulfilling action, and mental representation of that which may be presented linguistically. JH is missing Schnädelbach’s point of criticism (which I’ve amplified, of course, turning the point to my interest—but my present point relies on nothing other that JH’s quotation of Schnädelbach).

Though he’s missing Schnädelbach’s point (and mine—bad, bad, JH), I grant that:

JH: the propositional structure of knowledge, in the teleological structure of action, and in the communicative structure of speech, we come upon various roots of rationality. (309)

But we have to see that it is various roots of reason that we come upon: the epistemic, purposeful, and communicative interest of intelligent actors.

Valuation can be parsed into epistemic, purposeful, and communicative components or modes. But valuation can be parsed in other ways, too! (Putnam refers to the entirety of communicative rationality as a norm, in the sense of a basic value among other values). Other ways lose plausibility in JH’s sociocentric point of view (which tends to marginalize contrary views in terms of subjectivism, “subject-centered reason,” and “philosophy of consciousness”), but other ways do have plausibility (I would argue) relative to a richer sense of reason than JH allows—a richer sense that better accords with the normal sense of reason in society (lexically) and in cognitive science than does JH’s.

These [roots of rationality] do not for their part appear to have common roots, at least not in the discursive structure of justificatory practices,... (ibid.)

I’m not surprised. [March 2014: Confine rationality to justificatory sense, and you get “their part.”] Yet, discursive reflection [can] provide the appropriate venue for testing this claim (a later posting [which didn’t happen]).

...nor in the reflexive structure of the self-relation of a subject participating in discourses.

Again, discursive reflection may provide the way. [March 2014: I had in mind a difficult discursive venture that I didn’t make time to do.]

It is more probably the case that the structure of discourse establishes an interrelation among the entwined structures of rationality (the structures of knowledge, action, and speech)....

No, the structuring of knowledge, intelligent acting, and insightful articulating are keystones (among others) of reasoning. The entwining of capabilities provide a basis for reflexively establishing a discursive interrelation among emergent structures. [March 2014: This “entwining” might show as the multimodality of inquiry, a notion that involves a holism of ontogenic backgrounding. My implicit anticipation was that discourse is a kind of originative inquiry, not primarily an exercise in high-scale justification.] It’s not “the structure of discourse” that establishes, rather reflexivity inquiring into itself discursively (one’s comprehension of discursivity belongs to the degree or level of reflexivity that gains that interrelatedness clearly).

But putting the matter this way is merely to be working with JH’s vocabulary, saying what I can as a transformation of his idiom., in a sense, bringing together the propositional, teleological, and communicative roots.

So, here we are: on the verge of bringing together, in discursive thinking (which JH calls “discursive rationality” in the next section of “Some Further Clarifications...”).

According to such a model of intermeshed core structures....

...though “intermeshed” might now be a numinous notion...

...discursive rationality owes its special position not to its foundational but to its integrative role. (ibid.)

OK, I’ll take JH at his word on what discursive rationality can only do. But I will show that discursive reflexivity can be another matter.

Here we have a potential differentiation between discursive integration and discursive constructibility or individuation. A difference between individuation and integration turns out to be vital as well for understanding identity formation in general (I would argue, but that’s a long story).

...discursive rationality...emerges out of communicative rationality. (309)

...even though (now, get this)...

Communicative rationality does not constitute the overarching structure of rationality [vis-à-vis epistemic and teleological rationality, which “communicative rationality remains on a level with”] but rather [communicative rationality is] one of three core structures that are, however, interwoven with one another by way of the discursive rationality that emerges out of communicative rationality.

Isn’t that fun? Well, sure, for JH, discursive rationality is the idealized reflexivity of communicative rationality. Communicative rationality itself is interwoven with other modes of rationality by way of what emerges from itself.

But how can C be on the same “level” as T and E such that their “bringing together,” their “intermesh,” their interweaving emerges from C? Equiprimordial kinds (only one of which is dialogal) are likened to dialogue partners (which looks like a personification of action types). For JH, it’s not merely that three kinds of action become intermeshed in dialogal discourse (which puts communicative action on a different level); other kinds of action are conceived as subjects in communication.

Thus, we are tacitly invited (delightful to me) to understand the subjectivity of epistemic (theoretical) reason and the subjectivity of purposeful (practical) reason as an internal intersubjectivity of reason in itself, a reflexivity that is intersubjective. Yes, indeed, that’s Habermasian thinking!

So, here we have an intersubjectivity whose intersubjectivity is communicative reason on the same level as the two starkly different kinds of “subjectivity” (personification of action types). This has to arise from a lifeworld background that conceives of belief formation’s subjectivity and action’s subjectivity as on the same level as communication. The interests in knowledge and action are on the same level as the interest in communication.

But what comes first? Logically, don’t you have to have elements in order to have a relation? It seems then that one has to have an interest in knowledge and action before there can be “communication” between them—a communication which can’t be literally like that between two persons, because interests and abilities aren’t persons. (You might say that, logically speaking, you can postulate a relation prior to any variables that may be associated with the relation in a well-formed formula, but that would be superfluous to the present context, unless you want to claim that the non-personified “communication” between belief formation and purposeful action arises from a communicative relation that’s prior to any really interpersonal communication, and you have to say what kind of communicative relation without elements that is. I would say, if anything, it’s intelligence, but the interplay of it’s components don’t work like a given language; this is easy to demonstrate, given that the typical components of a theory of intelligence don’t work like a grammar; I’ll return to this point briefly below.)

In fact, according to infant and early child development research, abilities for knowledge-acquisition and purposeful action precede the ability to communicate. JH’s metaphor won’t work for developmental studies. Above, I’ve sketched how reflexivity can’t be assimilated to JH’s sense of discursive rationalization. Now, the metaphor of communicating kinds in a discursive rationality begs the question of what reflexivity is, such that one’s thinking may do all kinds of problem finding, discovering, and realization.

In any case,...

This picture should not, however, be misunderstod in a mentalist way.

Nor should it be understood in a socio-rationalistic way. Yet, JH’s “picture” isn’t what we basically should understand, since the issue of discursive reflexivity can’t be reduced to discursive rationality. (One might think that JH agrees, since he titles the next section of “Some Further Clarifications...” “Discursive Rationality and Reflection.” So, here we are, again, on the verge of “bringing together.”)

Discursivity fundamentally depends on one’s capability of thinking, which is a mental matter. JH’s proper aversion to mentalism is not relevant here, since mentalism is a theory of mind and an approach to the world which has no credence in contemporary cognitive science and philosophy of mind (which risks physicalism, not mentalism). So, I agree with JH about mentalism, but I think that there can importantly be realism in discursive reflexivity beyond some personification of action types.

JH is misleading, but typically true to form, when he comes to closure (I’ve been faced here with a mere 3 pages of “Some Further Clarifications...”!) by stating that:

Just as communicative rationality may not be equated with linguistically embodied rationality in general, epistemic and telelogical rationality are not of a prelinguistic nature. (309)

I suppose that JH’s point here is that communicative rationality contains epistemic and telelogical aspects, while the latter are linguistic by nature. Communicative rationality isn’t epistemic and teleological by nature, since linguistic nature inhabits everything; so, consequently, non-communicative action types have a linguistic nature.

In fact, epistemic and telelogical action aren’t fundamentally linguistic, since a full sense of intelligent capability belongs to all action types, and intelligence altogether can’t be assimilated to linguistic intelligence (which I’ve argued earlier, relative to the cognitive linguist Ray Jackendoff, Foundations of Language, Oxford 2002; and Howard Gardner’s well-corroborated theory of “multiple intelligences,” which R.J. Sternberg weaves into his own “triarchic” theory; see Metaphors of Mind, Cambridge UP 1990, ch. 11).

An accurate theory of reason must also include emotional intelligence (e.g., see neuroscientist A.R. Damasio, The Feeling of What Happens, Harcourt 1999).

Neither, then, can reflexive reason—manifold intelligence—be conceived as linguistic rationalization. Reflexivity (creative problem solving, inquirial discovery, etc.) may incorporate spatial, algebraic, musical, empathic, kinaesthetic, and self-realizing, as well as linguistic, modes (This is Gardner’s 7-fold sense of multiple intelligences), in discerning analogues, associations, complementarities, etc. which contribute to gaining insight.

But I don’t intend to fly all over the place in looking at JH’s sense of discursive reflexivity next (though I do like to fly). [March 2014: I believe I didn’t continue my irreverent venture.]


This is used in my discussion of Habermas’s “Norms and Values” essay on Putnam, March 2014, “...pertaining to capable intelligence...”.

Also: This discussion is associated with the “conceptual inquiry” area of

Be fair. © 2017, g. e. davis