Form, Substance

Faced with questions like what the world is “made of”, modern people have generally assumed that it must be some kind of “stuff”. The usual presumed answer is some sort of matter-stuff, or less commonly some sort of mind-stuff.

Plato and Aristotle already suggested a radical alternative to this way of thinking that takes the accent away from “stuff” altogether. Aristotle especially developed a rich account of how we think about these kinds of things, by looking at how we express these kinds of questions, and what we are implicitly trying to get at when we ask them.

The most obvious simple answer attributed to Plato and Aristotle is form. In Aristotle’s case, one should also mention what has been traditionally called substance.

Etymologically, eidos — the Greek word we translate as “form” — seems to begin from a notion of visual appearance, with an emphasis on shape. It acquired a more abstract sense related to geometrical figure. Plato attributed great significance to the practice of geometry as an especially clear and perspicuous kind of reasoning, but he also recognized a broader kind of reasoning associated with a dialectic of question and answer, which comes into play especially where questions of value are concerned. From a point of view of ethical practice and human life generally, questions about what something really is and why this rather than that are more important than what things are made of (see What and Why). Already with Plato, “form” came to be inseparable from meaning.

Aristotle’s classic discussion of substance (ousia) in the Metaphysics starts from the idea of a substrate in which properties inhere. This most superficial level later inspired the Greek grammarians to articulate the notion of a grammatical subject of predicates. In what I think is the single greatest example of ancient dialectic, Aristotle gradually steps back from the simple notion of a substrate. Substance becomes “what it was finally to have been” a thing, at the end of the day so to speak. But then this is further interrogated to disclose the level of actuality, or what is effectively operative in a process. But it turns out that actuality is not complete in itself. What is effectively operative does not form a self-contained whole that explains everything. A fuller understanding must take into account potentiality, which leads to a transition away from simple actuality to a larger perspective of processes and degrees of actualization, in which nothing is simply given just as it is. Aristotle was especially concerned with the forms of living things, which have this character.

The more interesting senses of form for Plato and Aristotle do not refer to something that could be simply given. In line with this but in a more speculative mode, Plotinus suggested that every form somehow in a way “contains” all other forms. The poetic truth in this is that the articulation of one form depends on the articulation of other forms, and while everything in some sense coheres, we have no unconditional starting point. We always begin in the middle somewhere, in a context that has yet to be fully elaborated. The work of elaboration is itself the answer. (See also Interpretation.)

Reference, Representation

The simplest notion of reference is a kind of literal or metaphorical pointing at things. This serves as a kind of indispensable shorthand in ordinary life, but the simplicity of metaphorical pointing is illusory. It tends to tacitly presuppose that we already know what it is that is being pointed at.

More complex kinds of reference involve the idea of representation. This is another notion that is indispensable in ordinary life.

Plato and Aristotle used notions of representation informally, but gave them no privileged status or special role with respect to knowledge. They were much more inclined to view knowledge, truth, and wisdom in terms of what is reasonable. Plato tended to view representation negatively as an inferior copy of something. (See Platonic Truth; Aristotelian Dialectic; Aristotelian Semantics.)

It was the Stoics who first gave representation a key role in the theory of knowledge. The Stoics coupled a physical account of the transmission of images — bridging optics and physiology — with very strong claims of realism, certain knowledge both sensory and rational, and completeness of their system of knowledge. In my view, the Stoic theory of representation is the classic version of the “correspondence” theory of truth. The correspondence theory treats truth as a simple “correspondence” to some reality that is supposed to be known beyond question. (Such a view is sometimes misattributed to Plato and Aristotle, but was actually quite alien to their way of thinking.)

In the Latin middle ages, Aquinas developed a notion of “perfect” representation, and Duns Scotus claimed that the most general criterion of being was representability. In the 17th century, Descartes and Locke built foundationalist theories of certain knowledge in which explicitly mental representations played the central role. Descartes also explicitly treated representation in terms of mathematical isomorphism, representing geometry with algebra.

Taking putatively realistic representational reference for granted is a prime example of what Kant called dogmatism. Kant suggested that rather than claiming certainty, we should take responsibility for our claims. From the time of Kant and Hegel, a multitude of philosophers have sharply criticized claims for certain foundations of representational truth.

In the 20th century, the sophisticated relational mathematics of model theory gave representation renewed prestige. Model-theoretic semantics, which explains meaning in terms of representation understood as relational reference, continues to dominate work in semantics today, though other approaches are also used, especially in the theory of programming languages. Model-theoretic semantics is said to be an extensional rather than intensional theory of meaning. (An extensional, enumerative emphasis tends to accompany an emphasis on representation. Plato, Aristotle, Kant, and Hegel on the other hand approached meaning in a mainly intensional way, in terms of concepts and reasons.)

Philosophical criticism of representationalist theories of knowledge also continued in the 20th century. Husserl’s phenomenological method involved suspending assumptions about reference. Wittgenstein criticized the notion of meaning as a picture. All the existentialists, structuralists, and their heirs rejected Cartesian/Lockean representationalism.

Near the end of the 20th century, Robert Brandom showed that it is possible to account very comprehensively for the various dimensions of reference and representation in terms of intensionally grounded, discursive material inference and normative doing, later wrapping this in an interpretation of Hegel’s ethical and genealogical theory of mutual recognition. This is not just yet another critique of representationalism, but an actual constructive account of an alternative, meticulously developed, that can explain how effects of reference and representation are constituted through engagement in normative discursive practices — how reference and representation have the kind of grip on us that they do, while actually being results of complex normative synthesis rather than simple primitives. (See also Normative Force.)

Phenomenological Reduction?

This is a follow-up to my earlier article on Husserlian and existential phenomenology in light of the past year’s reading of Paul Ricoeur. In The Conflict of Interpretations (French ed. 1969), Ricoeur discusses the impact of his own view of hermeneutics as a “long detour” essential to understanding.

Ricoeur wrote that “It is in spite of itself that [Husserlian] phenomenology discovers, in place of an idealist subject locked within its system of meanings, a living being which from all time has, as the horizon of all its intentions, a world, the world. In this way, we find delimited a field of meanings anterior to the constitution of a mathematized nature, such as we have represented it since Galileo, a field of meanings anterior to objectivity for a knowing subject. Before objectivity, there is the horizon of the world; before the subject of the theory of knowledge, there is operative life” (p. 9). “Of course, Husserl would not have accepted the idea of meaning as irreducibly nonunivocal” (p. 15).

“In truth, we do not know beforehand, but only afterward, although our desire to understand ourselves has alone guided this appropriation. Why is this so? Why is the self that guides the interpretation able to recover itself only as a result of the interpretation? …the celebrated Cartesian cogito, which grasps itself directly in the experience of doubt, is a truth as vain as it is invincible…. Reflection is blind intuition if it is not mediated by what Dilthey called the expressions in which life objectifies itself. Or, to use the language of Jean Nabert, reflection is nothing other than the appropriation of our act of existing by means of a critique applied to the works and the acts which are the signs of this act of existing…. [R]eflection must be doubly indirect: first, because existence is evinced only in the documents of life, but also because consciousness is first of all false consciousness, and it is always necessary to rise by a corrective critique from misunderstanding to understanding” (pp. 17-18). This is a nice expression of what I take to be one of the greatest lessons of Aristotle and Hegel (see First Principles Come Last; Aristotelian Actualization; What We Really Want.)

For Ricoeur, Husserlian phenomenological reduction ceases to be a “fantastic operation” identified with a “direct passage”, “at once and in one step”. Rather, “we will take the long detour of signs” (p. 257).

Husserl’s “reductions” reduced away reference to putatively existing objects in favor of a sole focus on what would be the Fregean sense in meaning. Ricoeur wants to reintroduce reference, and in this way to distinguish a semantics that includes consideration of reference from a semiology addressing pure sense articulated by pure difference. Reference for Ricoeur is not a primitive unexplained explainer, but something that needs to be explained, and a big part of the explanation goes through accounts of sense. Ricoeur also wants to connect reference back to the earlier mentioned “self that guides the interpretation”, which again functions as an end rather than being posited as actual from the outset.

Similarly to his critique of phenomenological reduction “at once and in one step”, he criticizes Heidegger’s “short route” that in one step simply replaces a neo-Kantian or Husserlian “epistemology of interpretation” with an “ontology of understanding”. Ricoeur is a lot more deferential to Heidegger than I would be at this point, but for Ricoeur such an ontology is again only a guiding aim, and not a claimed achievement like it was for Heidegger. I think this makes Ricoeur’s “ontological” interest reconcilable with my own “anti-ontological” turn of recent years, because my objections have to do with claimed achievements. I broadly associate Ricoeur’s modest ontology-as-aim with my own acceptance of a kind of inquiry about beings that avoids strong ontological claims. Even Heidegger emphasized Being as a question.

Ricoeur of course rejects foundationalist epistemology (see also Kant and Foundationalism), but sees both an epistemology of interpretation and an ontology of understanding as aims guiding the long detour. He effectively contrasts the long path of investigation of meaning with the short path of appeals to consciousness (see also Meaning, Consciousness).

I actually like the idea he attributes to Husserl of reducing being to meaning or the sense(s) of being. If meaning is fundamentally nonunivocal as Ricoeur says rather than univocal as Husserl wanted, this would not be idealist in a bad sense.

Brandom’s simpler suggestion that reference is something real but that it should be ultimately explained in terms of sense seems to me a further improvement over Ricoeur’s apparent notion of reference as a kind of supplement to sense that nonetheless also needs to be explained in terms of sense, but without being reduced to it. I see the inherently overflowing, non-self-contained nature of real as compared to idealized being/meaning as making a supplement superfluous. (See also Reference, Representation; Meant Realities.)


“Imagination” is said in at least three major ways.  Aristotle minimalistically characterized phantasia as a production of images that both plays a role in our experience of sense perception and can operate independent of it, as in dreaming.  Spinoza treated imagination as kind of a passive belief.  For him, this was strongly associated with common illusions and wishful thinking – especially with regard to our status as agents — in ordinary life.  The Romantics identified imagination with creativity.

Beatrice Longuenesse in her marvelous Kant and the Capacity to Judge has developed in detail Kant’s argument that the same basic “categories” used in reflective thought are already implicit in our pre-reflective apprehensions of things in what Kant called a synthesis of imagination.  I think this means not that the Kantian categories have some pre-given or metaphysical status, but rather that for the kind of beings we are, even “pre-reflective” apprehensions have some dependency on previous reflective apprehensions.  We are never either entirely active or entirely passive.  (See also Passive Synthesis, Active Sense; Voluntary Action; Middle Part of the Soul.)

Richard Kearney in On Paul Ricoeur: The Owl of Minerva nicely develops Ricoeur’s view that imagination is not so much a special way of seeing as “the capacity for letting new worlds shape our understanding of ourselves…. This power would not be conveyed by images, but by the emergent meanings in our language” (quoted in Kearney, p. 35).  According to Kearney, Ricoeur associated imagination first and foremost with “semantic innovation”.  What Aristotle in a different context called “searching for a middle term” is an aspect of this creativity with respect to meaning.

The Greek root for “poetry” (poiesis) fundamentally means making or doing in a much more general sense.  The Romantics added a stress on innovation, which they saw as coming from the inner depths of the soul.  Ricoeur’s treatment of imagination as fundamentally involving the emergence of new meaning nicely takes up the Romantic stress on imagination as innovation, without depending on the Romantics’ dubious metaphysical psychology of interiority.  (See also Personhood; Reason, Nature.)

Brandom on Reason

In the introduction to Reason and Philosophy (2009), Brandom identifies with “a venerable tradition that distinguishes us as rational animals, and philosophy by its concern to understand, articulate, and explain the notion of reason….  Kant and Hegel showed us a way forward for a rationalism that is not objectionably Cartesian, intellectualist, or anti- (or super-) naturalist.  Nor need it treat the ‘light of reason’ as unacquired or innate” (pp. 1-2; emphasis in original throughout).

“Rational beings are ones that ought to have reasons for what they do, and ought to act as they have reason to” (p.3).

“Taking something to be subject to appraisals of its reasons, holding it rationally responsible, is treating it as someone: as one of us (rational beings).  This normative attitude toward others is recognition, in the sense of Hegel’s central notion of Anerrkennung” (p. 3).

The role of recognition makes things like authority and responsibility into social statuses.  These “are in principle unintelligible apart from consideration of the practical attitudes of those who hold each other responsible, acknowledge each other’s authority, attribute commitments and entitlements to each other” (pp. 3-4).

If we take meaning seriously, we cannot take it for granted.  Inferential articulation is involved not only in determining what is true, but also in the understanding of meanings.  What we mean and what we believe are actually interdependent.  He refers to Wilfrid Sellars’ thesis that no description can be understood apart from the “space of implications” in which the terminology used in the description is embedded.  “Discursive activity, applying concepts paradigmatically in describing how things are, is inseparable from the inferential activity of giving and asking for reasons” (p. 8).  

“[T]he acts or statuses that are givings of reasons and for which reasons are given – are judgings, claimings, assertings, or believings.  They are the undertakings or acknowledgements of commitments” (p. 9).  “[R]ationality is a normative concept.  The space of reasons is a normative space” (p. 12).  Philosophy should be concerned not just with pure logic and semantics, but with “the acknowledgement and attribution of… statuses such as responsibility and authority, commitment and entitlement” (p. 13).

Dialectic Bootstraps Itself

Here is a subtle but vital point. I’ve just reiterated that dialectic assumes no prior truth. Dialectic approaches coherence in an iterative and incremental way, sometimes backing up and trying a new path. As a philosopher in what I take to be the genuine spirit of Plato and Aristotle, I think seriously taking up such a work is the very best we mortals can honestly do to achieve high levels of practical confidence about the things that matter most in life. No, it does not have the precision or definitiveness of mathematics, but mathematics is like Aristotelian demonstration in that all it tells us with certainty is that certain conclusions follow from certain premises, so the conclusions are only as applicable to life as the premises and their assumed mapping to the real world.

The vital point is that dialectic — the development of richer meaning — can make real progress, without ever assuming a foundation. No, we’ll never say the last word, but we can iteratively and incrementally build cumulative results. With iterative and incremental development of coherent articulations of what we care about and an openness to acknowledging error, we can progressively improve interpretive confidence.

I think this is what is behind Hegel’s somewhat mystifying talk about spirit producing itself. (See also Interpretation; Dialogue; Ethical Reason; Practical Reason; The Autonomy of Reason; Openness of Reason; Reason, Feeling.)

Narrated Time

The third volume of Paul Ricoeur’s Time and Narrative returns to more overtly philosophical themes. From the beginning of volume 1, he has been using Augustine’s aporias concerning time as a sort of background to everything else he considers. He had suggested that both Husserl’s phenomenology of internal time consciousness and Heidegger’s existential phenomenology of time — contrary to the intentions of their authors — ended up in aporias similar to Augustine’s.

Ricoeur says that neither a phenomenology of the experience of time nor a “cosmological”, measurement-oriented approach to its objective aspects can avoid some dependency on the other. “The distension of the soul alone cannot produce the extension of time; the dynamism of movement alone cannot generate the dialectic of the threefold present” (vol. 3, p. 21). His hope is that a poetics of narrative, even if it too is unable to resolve the Augustinian aporias, can at least make them “work for us” (p. 4). It will develop the “complicity as well as the contrast” (p. 22) between the two approaches.

He will subordinate the “dimension of reference to the hermeneutic dimension of refiguration” developed in volume 2’s discussion of literary narrative (p. 5, emphasis added). The approach to a “real” historical past can then be understood in terms of a narrative refiguration, rather than vice versa. (This differs significantly from Brandom’s subordination of reference to what might be called a hermeneutic dimension of material inference, but both Ricoeur and Brandom are putting some kind of hermeneutics or interpretation of meaning conceptually ahead of reference to “things” (see also What and Why; Objectivity of Objects). For both of them, reference is still a valid concept, but it is something that potentially stands in need of explanation, rather than something that provides an explanation.)

For Ricoeur here, “pure” semantics and foundationalist epistemology are both superseded by a “hermeneutic of the ‘real’ and the ‘unreal'” (p. 6) he expects will yield insight into both history and fiction. History and fiction are two ultimately interdependent modes of narrative refiguration, so that there is no history without an element of creative fiction, but also no fiction without an element of something like what is involved in historical reconstruction. This seems like an important and valid point.

The main body of this volume contains further elaboration on various matters he discussed before (see Time and Narrative; Ricoeur on Augustine on Time; Emplotment, Mimesis; Combining Time and Narrative; Ricoeur on Historiography; Literary Narrative; Narrative Time). In separate posts, I will selectively comment on a few parts of this. (See Philosophy of History?; Ricoeur on Foucault.)

At the end, he wants to “verify at what point the interweaving of the referential intentions of history and fiction constitutes an adequate response” (p. 242) to the aporia resulting from the interdependence of the phenomenological and “cosmological” views of time. Second, there is the question “what meaning to give to the process of totalization of the ecstases of time, in virtue of which time is always spoken of in the singular” (ibid). He expects the answer here to be less adequate. It will yield “a premonition of the limits ultimately encountered by our ambition of saturating the aporetics of time with the poetics of narrative” (p. 243). Finally, most “embarrassing” of all is the new question, “can we still give a narrative equivalent to the strange temporal situation that makes us say that everything — ourselves included — is in time, not in the sense given this ‘in’ by some ‘ordinary’ acceptation as Heidegger would have it in Being and Time, but in the sense that myths say that time encompasses us with its vastness” (ibid; emphasis in original). This, he suggests, ultimately remains a mystery in the Marcelian sense.

“Narrated time is like a bridge set over the breach constantly opened up by speculation between phenomenological time and cosmological time” (p. 244). “Augustine has no other resources when it comes to the cosmological doctrines than to oppose to them the time of a mind that distends itself” (ibid), but his meditations on Creation implicitly require a “cosmological” time. Aristotle’s cosmological view made time dependent on motion but distinct from it as a measurement of motion, but any actual measurement seems also to depend on some action performed by a soul.

Husserl elaborated a view of something like Augustine’s threefold present, which included memory and anticipation as well as current attention. He spoke of time as constituted through “retention”, “protention”, and a sort of comet-like duration rather than a point-like present. I am barely skimming the surface of a sophisticated development.

Ricoeur was a great admirer of Husserl, but in this case suggests that Husserl failed to achieve his further goal of establishing the primacy of phenomenological time with respect to other sorts of time. For Husserl, Ricoeur says, the constitution of phenomenological time depends on a “pure hyletics of consciousness” (ibid), but any discourse about the hyletic (i.e., relationally “material” in a broadly Aristotelian and more specifically Husserlian sense) will depend on the “borrowings it makes from the determinations of constituted time” (ibid). Thus for Ricoeur, the articulation of what Husserl wanted to be a purely constitutive phenomenological time actually depends on what he had wanted to treat as constituted results. As a consequence, despite Husserl’s wishes, phenomenological time should not be simply said to be purely constitutive.

Heidegger’s “authentic temporality” takes this aporia to its “highest degree of virulence” (p. 245). Being-in-the world does appear as a being-in-time. However, that time remains resolutely “individual in every case” (ibid), owing to Heidegger’s fixation on being-toward-death.

For Ricoeur, the “fragile offshoot” of the “dialectic of interweaving” of the “crisscrossing processes of a fictionalization of history and a historization of narrative” (p. 246) is a new concept of narrative identity of persons and communities as a practical category. A narrative provides the basis for the permanence of a proper name. He alludes to a saying of Hannah Arendt that to answer the question “who” is to tell the story of a life. Without such a recourse to narration, there is an “antinomy with no solution” between simply positing a substantially or formally identical subject, or with Hume and Nietzsche globally rejecting such a subject as an illusion.

(I have proposed a different “middle path” by decoupling actual subjectivity from the assumption of more unity than can be shown, and associating it more with what we care about and hold to be true than with “us” per se. Yet I also find myself wanting to tell a story of how the sapient “I” and the sentient “me” I want to distinguish are nonetheless interwoven in life, which is also what I think Hegel wanted to do.)

Ricoeur here introduces the connection between narrative identity and an ethical aim of “self-constancy” that he developed later in Oneself as Another. “[T]he self of self-knowledge is not the egotistical and narcissistic ego whose hypocrisy and naivete the hermeneutics of suspicion have denounced, along with its aspects of an ideological superstructure and infantile and neurotic archaism. The self of self-knowledge is the fruit of an examined life, to recall Socrates’ phrase” (p. 247). In Hegelian terms, this is the “self” involved in self-consciousness. In my terms, who we are is defined by what we care about and how we act on that. It is a “living” end or work always in progress rather than an achieved actuality. For Ricoeur, psychoanalysis and historiography provide “laboratories” for philosophical inquiry into narrative identity.

The second aporia concerned the unity or “totality” of time. “The major discovery with which we have credited Husserl, the constitution of an extended present by the continuous addition of retentions and protentions… only partially answers this question” (p. 252). It only results in partial “totalities”.

(While rejecting claims to unconditional “totality”, Ricoeur here accepts the terminology of “totalization” as an aim. I prefer to take something like Ricoeur’s conclusion that we only ever achieve partial “totalities” as a ground for saying that even as an aim, we should speak more modestly of (always partial and local) synthesis rather than “totalization”. To my ear, even as an aim “totalization” sounds too univocal and predetermined. I want to say that an aim of totalization is inherently unrealizable for a rational animal, whereas any end we turn out to have been actually pursuing based on interpretation of our actions must in some sense have been realizable. Ricoeur emphasizes that the mediation involved is imperfect, so that I think the difference in his case is merely verbal, but other authors’ use of “totalization” is more problematic.)

For Ricoeur, “the constitution of a common time will then depend on intersubjectivity” (p. 253), rather than on the unity of a consciousness. We should replace a “monological” theme of fallenness with a “dialogical” theme of being affected by history. Meanwhile, all initiative is in a sense “untimely”. The dialogical character of a historical present opens onto the same space of reciprocity as the making of promises. “[T]he imperfect mediation of historical consciousness responds to the multiform unity of temporality” (p. 257). But meanwhile, “it is not certain that repetition satisfies the prerequisites of time considered as a collective singular” (ibid). In this context he speaks of “an original status for the practical category that stands over against the axiom of the oneness of time” (ibid), and of a return to Kantian practical reason that “can be made only after a necessary detour through Hegel” (p. 258).

Ricoeur proposes “an epic conception of humanity”. Nonetheless, the “good correlation between the multiform unity of the ecstases of time and the imperfect mediation of the historical consciousness” (p. 259) cannot be attributed to narrative. “[T]he notion of plot gives preference to the plural at the expense of the collective singular in the refiguration of time. There is no plot of all plots capable of equaling the idea of one humanity and one history” (ibid). Narrativity does not so much resolve the aporias of time as put them to work.

The final aporia concerned the inscrutability of a unified time. He associates this with our non-mastery of meaning. We are “pulled back” toward an archaic, mythical, poetic form of thinking of the oneness of time that “points toward a region where the claim of a transcendental subject (in whatever form) to constitute meaning no longer holds sway” (p. 263). About returning to an origin we can only speak in metaphors. Finally, he asks if it is possible to speak of a narrative refiguring of the unrepresentability of time. “It is in the way narrativity is carried toward its limits that the secret of its reply to the inscrutability of time lies” (p. 270). “[T]he narrative genre itself overflows into other genres of discourse” (p. 271). Fiction multiplies our experiences of eternity in various ways, “thereby bringing narrative in different ways to its own limits” (ibid). It serves as a laboratory for an unlimited number of thought experiments. It “can allow itself a certain degree of intoxication” (ibid).

“It is not true that the confession of the limits of narrative abolishes the idea of the positing of the unity of history, with its ethical and political implications. Rather it calls for this idea…. The mystery of time is not equivalent to a prohibition directed against language. Rather it gives rise to the exigence to think more and to speak differently” (p. 274). It is only in the context of a search for narrative identity “that the aporetics of time and the poetics of narrative correspond to each other in a sufficient way” (ibid).

Literary Narrative

Resuming the thread on Ricoeur’s Time and Narrative, the beginning of volume 2 makes it clear he thinks the stakes in discussing fictional narrative — distinguished from historical narrative by not being oriented toward anything analogous to historical truth — extend well beyond the traditional concerns of literary criticism. (Of course, going beyond those traditional concerns is pretty much the norm for literary critics these days.) Let me say that I am no scholar of literature, though I used to write experimental “language on language” poetry, and have read a bit of literary theory mainly for its philosophical content.

Philosophically, he will defend the “precedence of our narrative understanding in the epistemological order” (vol. 2, p. 7). There will be a great “three-way debate” between lived experience, historical time, and fictional time. He will extensively take up the idea of the “world” projected by a text. There will be a sort of analogy between the place of a structuralist or semiotic analysis of a literary text and the work of explanation in history, but Ricoeur says the analysis will eventually show that literary narrative has different relations to time than historical narrative.

Thinking about literary narrative, he suggests, will turn out to be more helpful in shedding light on Augustine’s paradoxes of the experience of time. In the literary case, there will be a distinction between the time of the act of narration and the time of the things narrated.

The continuing relevance of a notion of plot in modern literature, he says, needs to be shown rather than assumed. Rather than extending and further abstracting the general principle of formal composition that Aristotle had begun to articulate, modern literary studies began with an odd combination of struggle against old conventions, concern for increased realism, and inappropriate borrowing from models of ancient genres. According to Ricoeur, this resulted in a mutilated, dogmatic, not very interesting notion of plot associated mainly with a linear sequence of events, that “could only be conceived as an easily readable form” (p. 9).

Modern novels, on the other hand, have greatly increased the importance of character, and explored the dynamic process of its growth. They may take alternate forms, such as streams of consciousness, diaries, or exchanges of letters. Action accordingly has to be understood in a broader and deeper way no longer limited to external behavior, encompassing things like growth of character and moments in a stream of consciousness. Once this is done, he says, a generalized notion of imitation of action will again apply.

But the early English novelists “shared with empiricist philosophers of language from Locke to Reid” (p. 11) an ideal of purely representational language stripped of metaphor and figurative constructs. “Implicit in this project is the reduction of mimesis to imitation, in the sense of making a copy, a sense totally foreign to Aristotle’s Poetics” (p. 12). I’m more inclined to think so-called literal language is just a limit case of metaphor. I would also note that this representationalist paradigm of transparency is the direct opposite of the “language on language” perspective. “Today it is said that only a novel without plot or characters or any discernible temporal organization is more genuinely faithful to experience” (p. 13). The kind of justification offered, he says, is the same as the one for naturalistic literature — reproduction of experience rather than synthesis. I’m sure someone must have done that, but appeals to experience are somewhat inimical to the structuralist “language on language” view.

Ricoeur wants to suggest that literary paradigms originate “at the level of the schematism of the narrative understanding rather than at the level of semiotic rationality” (p. 15). He will argue that a purely semiotic approach to narrative has the same weaknesses as the positivist “covering law” model in history. I’m a little confused by this, because earlier he more charitably compared the semiotic approach to historical explanation in general, which he had presented as legitimately different from narrative understanding.

He suggests that structural analysis at lower levels like phonology or lexical semantics does not lose nearly as much context of meaning as structural analysis of narrative, which aims to reduce away all temporal elements and replace them with logical relationships. He also says the identity of a style is transhistorical, not atemporal, and that styles are perennial rather than eternal.

He contests the assertion of Roland Barthes that there is an “identity” between language and literature, and that each sentence already has the essential features of a narrative. Behind what Ricoeur is objecting to, I think, may be the additional idea of a strictly compositional, bottom-up interpretation of meaning, which he alluded to earlier. In a formal context like that of structural analysis, compositionality is an extremely important property, but in a broader hermeneutics, I agree with Ricoeur that a bottom-up approach is basically a non-starter. My own past enthusiasm for structuralism had much more to do with its relational, difference-before-identity aspect. I’ve always had severe doubts about any bottom-up reduction when it comes to meaning.

On the other hand, while recognizing many valuable contributions of Husserl and his followers, I fundamentally disagree with what I take to be that tradition’s identification of “consciousness” with what I take Hegel to have sharply distinguished from consciousness as “self-consciousness”.

On my reading, this distinction is the radical “break” in Hegel’s Phenomenology. I have glossed “self-consciousness” as actually other-focused even though it does involve a unity of apperception, and as anything but a species of a genus “consciousness”. I think unity of apperception and Hegelian “self-consciousness” have to do with Aristotelian ethos and what we care about, as a discursive stance in relation to others. I do of course agree that there is consciousness, and that it has a kind of interiority of its own. I also agree that one of its features is something like Husserl’s “living present”, but I think consciousness and the living present belong to what Brandom calls our sentience rather than to what he calls our sapience, which I associate with unity of apperception and “self-consciousness”.

What attitude one takes on this question of the identity or distinctness of consciousness and self-consciousness matters greatly when it comes to something like the debate between structuralism and phenomenology in the tradition of Husserl. I care about rich concepts of reason and feeling, but “consciousness” not so much. I am not worried about the impact of structuralism on the living present, because I see them as pertaining to disjunct domains.

On the other hand, I think Ricoeur is right to be very doubtful whether narrative can be adequately understood without temporality, and right again to reject bottom-up determination of meaning. I am inclined to be sympathetic to his view that temporality cannot be reduced to logic.

I also think it makes a big difference whether one is considering a stereotyped form like the folk tales whose analysis by Vladimir Propp he discusses, or something as complex as a modern novel. Structural analysis may come much closer to yielding comprehensive insight in the one case, while falling much further short in the other. (Ricoeur is not satisfied even in the case of the folk tale.)

Ricoeur also discusses work on a higher level “logic of narrative” concerned with roles of characters, and the narrative grammar of Greimas. He says narrative has more to do with history than with logic. He makes the very valid point that the structuralist notion of the “diachronic” captures only a simple notion of succession. I agree that one should not look to structuralism for rich analyses of time itself, but that does not mean it cannot give us insight about things happening in time, which in turn does not mean it gives us the whole story either.

Rule of Metaphor

In The Rule of Metaphor, which contains essays from the early 1970s, Ricoeur aimed among other things to refute Frege’s apparent claim that poetry has no reference or denotation, but only a sense or connotation. According to Ricoeur, poetic language achieves a kind of “second-level reference” by suspending first-level reference. I tend to think all reference presupposes higher-order constructs, so I am sympathetic. This is also an argument for the general importance of metaphor.

In an appendix, he describes how the rise of structuralism in the 1960s — of which he remained critical — led him from a kind of existential phenomenology to a much closer engagement with language. His earlier emphasis on symbols gave way to a more general approach to hermeneutics, and he began to also engage with analytic philosophy.

At the beginning of the book, he notes how the study of rhetoric became much narrower after Aristotle, losing its connection with dialectic and philosophy. Later, he goes on to argue that meanings of sentences take precedence meanings of words, and meanings of whole discourses take precedence over meanings of sentences.

Apparently, some structuralist writers on rhetoric argued for the contrary, bottom-up approach starting with meanings of words. My own past interest in so-called structuralism never led in this direction; I was initially more concerned with the priority of relations over “things”, and later with the explanatory power of Foucaultian “discursive regularities”. I do think a compositional, bottom-up approach has great value in formal contexts, and that formal analysis is not irrelevant to ordinary language, but I think ordinary language meaning is best approached mainly in terms of material inference, which has a holistic character.