Contemporary Non-conceptualism, Conceptual Inclusivism, and the Yogācāra View of Language Use as Skillful Action

This essay seeks to explain the somewhat paradoxical Yogācāra Buddhist view of the possibility of non-conceptual language use (under what is called 'subsequent awareness', tatpṛṣṭhalabdhajñāna). The essay unpacks the school's understanding of non-conceptuality, drawing extensively on contemporary non-conceptualist theory, as well as on conceptualist accounts that are referred to here as "inclusivist." It is argued that the inclusivists' proposal that we view conceptual articulation itself as a form of practical-skillful action presents some intriguing affinities (but also important disaffinities) with the Yogācāra conception of language use, and takes us some way toward understanding what a concept is for a Yogācāra Buddhist.

Introduction

According to the early Yogācāra, following non-conceptual awareness (nirvikalpajñāna), the advanced bodhisattva is said to attain a state characterized by a "subsequent awareness" (tatpr.ṣṭhalabdhajñāna). Yogācāra thinkers identify this state with ultimate knowledge of causality and view it as involving a unique kind of conceptual activity and propositional attitudes, which are very different, however, from ordinary conceptual awareness insofar as they do not involve vikalpa. Translated back into the terms of some version of the contemporary debate between conceptualists and nonconceptualists, this would amount to something like the claim that we can use concepts in a non-conceptual way. But how can we make sense of that? Resorting to soteriology, which is sometimes the last refuge of a bemused philosopher, will not do here. Explaining away this seemingly paradoxical Yogācāra position by suggesting that under the subsequent awareness the bodhisattva uses concepts only conventionally, and without reifying them (a trick often pulled in the case of the Madhyamaka), leaves too much unaccounted for. It does not explain, for instance, how meaning is derived and publicly communicated under this framework, nor does it explain the normativity that is at play in conceptual articulation.

For making sense of the Yogācāra notion of non-conceptuality, what is required here, instead, is an explication of the school's theory of both meaning and linguistic performance. Elsewhere I have written extensively on¹ the former issue, and in the present article I wish to focus on the latter, examining what the Yogācāra view of conceptual activity and language use, under the subsequent awareness, entails for our understanding of linguistic performance.

In the first section, by way of introduction, I present the Yogācāra understanding of subsequent awareness based on arguments (in I.1) that I have previously made elsewhere,² and broadly discuss the compatibility of this framework with some of the presuppositions and terminology underlying contemporary non-conceptualist theories (in I.2).

Building on this ground, in the second section I proceed to unpack in greater detail the Yogācāra understanding of linguistic performance, drawing extensively on contemporary non-conceptualist theory (primarily Adrian Cussins in II.1),³ as well as on conceptualist accounts that I describe as "inclusivist" (mostly by Alva Noë and Joseph Rouse, in II.2.1 and II.2.2, respectively).⁴ I argue that the inclusivist proposal that we view conceptual articulation itself as a form of practical-skillful action presents some intriguing affinities (but also important disaffinities) with the Yogācāra conception of language use. Taking a close look at the overlap and the differences between these accounts helps to clarify the broader question of whether and how the contemporary distinction between non-conceptual and conceptual content aligns with the Yogācāra understanding of these terms, and takes us some way toward understanding what a concept is for a Yogācāra Buddhist.

I. The Yogācāra Notion of Subsequent Awareness

I.1. A Summary of the Yogācāra Notion of Subsequent Awareness

One of the concerns that underlie the Yogācāra discussion of the bodhisattva's non-conceptual (nirvikalpa) experiences is how to reconcile these experiences with their need to operate and communicate within saṃsāra. To address this concern, as mentioned, various Yogācāra sources adopt a division of labor between two types of knowledge: non-conceptual awareness (nirvikalpajñāna) and the "awareness obtained subsequent to it" (tatpr.ṣṭhalabdhajñāna), the latter said to be obtained by the advanced bodhisattva following non-conceptual awareness, as its name implies. Yogācāra thinkers appear to identify the latter awareness as the state in which ultimate knowledge of causality is attained. In the Triṃs'ikā-bhāṣya, for instance, Sthiramati explicitly correlates this awareness with the understanding of the dependent-nature as the causal interconnectedness of all essenceless phenomena.⁵

As such, this awareness appears to enable propositional knowledge and involve a unique kind of conceptual activity, which is very different, however, from ordinary conceptual awareness, and which allows the bodhisattva to remain, operate, and communicate effectively in the phenomenal world.

The understanding of the "subsequent awareness" as, foremost, a knowledge of causality is revisited in other Yogācāra sources, most notably the Mahāyānasaṃgraha (MS). Consider, for instance, the following reference to the subsequent awareness in the third chapter of the work:

This subsequent cognition (pr. ṣṭalabdhajñāna), which considers every creation (prabhava) coming from the store-consciousness and every object of concept (vijñaptinimitta) as a magic show (māyā), etc., is, in its essence, free of errors (prakṛtyaviparīta). Thus, in the same way that the magician (māyākāra) is free of doubt about the things produced by magic (māyākṛtadharma), so this bodhisattva is always unmistaken (viparyāsa) when he speaks of cause (hetu) and result (phala).⁶

Elsewhere,⁷ as mentioned, I have presented a detailed survey of the MS account of this topic, so here I will just recap by briefly indicating some key points. To this passage's understanding of subsequent awareness as the knowledge of phenomena in causal terms, the Upanibandhana (U)⁸ commentary adds that this knowledge—while free of concepts (nirvikalpa), pure and non-dual—is seen as the bodhisattva's "generator of speech" (vāksamutthāpaka) within the conventional order.⁹ Here the perspective of subsequent awareness seems to be identified with the conventional and communicative order—which is, however, pure and unerring with respect to causality, and free of the defilements—within which the bodhisattva operates. Furthermore, other sections in the MS tell us that in the absence of subsequent awareness, non-conceptual experience in itself is pointless (likened by the treatise to a mute who cannot communicate what he knows or a fool who fails to make sense of what he knows), if it is not eventually utilized for the salvific activity of the bodhisattva, which requires verbal communication and interaction.¹⁰

It should be noted, however, that according to the MS and its commentaries, while non-conceptual awareness is not intentional, it is not altogether ineffective, and allows the Bodhisattva to continue acting, but under a different conception of agency. Using several analogies to illustrate this point,¹¹ the text emphasizes the spontaneous and nonvolitional character of the Bodhisattva's salvific activity within this state, and explicitly states that such actions, while efficacious, do not entail agency in the sense of deliberation or intentional cognitive content.¹²

The notion of Buddhahood that emerges from this account of nonconceptual awareness is very similar to what Mario D'amato, following John Dunne (1996), has dubbed the model of the "mindless Buddha," according to which, behind appearances, a Buddha is seen to be ultimately without any cognitive conceptual content and without any utterances.¹³ This conception, however, is supplemented in the MS with an entirely different kind of understanding of the kind of efficacy and agency involved in the activity of the enlightened bodhisattva under the state of subsequent awareness. In this state one is aware (in retrospect) of having had a nonconceptual experience, commits this awareness to memory, and communicates it to others, culminating in a kind of knowledge that is said to lead to no less than the understanding of the sameness of all phenomena and to the ensuing transformation of the basis (ās'raya-parivrṭti). In this sense this state is also characterized by what is described as the "prosperity" or "success" ('byor pa, samṛddhi*)¹⁴ of the bodhisattva with respect to transforming reality—for instance from earth to gold—according to his aspirations (ās'aya).

Within the Yogācāra lore, ās'aya (used interchangeably with closely related terms, such as adhimukti) refers generally to the way in which volitional and intentional content affects and fashion one's perceptions of phenomenal appearances and existents.¹⁵ However, in the hands of the advanced bodhisattva, whose perception is no longer constrained by conceptual discrimination (and the allegedly objective givenness of phenomena), this term indicates the way in which he is able to manipulate his perception of phenomena at will. This very ability—by virtue of which different perceivers may experience the same object differently—serves the MS to ground a practical argument for the unreality of external objects as a source for experience.¹⁶ If non-conceptual awareness and what it entails is to be accepted as an experiential fact (as Yogācāra in fact accepts), then external objects simply cannot be thought to exist.

Let us recapitulate so far: the Yogācāra sources identify the ultimate knowledge of causal relations with the bodhisattva's state of subsequent awareness. While non-conceptual awareness is conceived of by Yogācāra as a core experience of the advanced bodhisattva, its characterization as ineffable and free of constructions by definition precludes any kind of intentional content or the possibility of communication. Within this framework, efficacious actions, while conducive to liberation, appear as a kind of unvolitional and spontaneous activation rather than a volitional reaction. For an enlightened bodhisattva, however, fulfilling her salvific role within saṃsāra—namely, acting and communicating effectively within the phenomenal world—requires an additional kind of subsequent awareness, which is characterized on the one hand as mundane (with a cognitive world that involves reflexivity, memory, verbal representation, explicit intention, and deliberation), and on the other hand as pure and free of constructions. This paradoxical characterization of the subsequent awareness, however, should be understood not as plainly incoherent but rather as emphasizing the way in which it stands in relation to and is indeed an extension of the preceding non-conceptual awareness.

Within the career of the bodhisattva, subsequent awareness appears to serve as the norm rather than the exception. It is not seen as a fall from (non-conceptual) grace but as the continuation and application of this state's insights in the phenomenal world. The knowledge of causality and the use of conceptualization that it presents are therefore not ordinary, but radically transformed. Insofar as it stands for an exhaustive perspective of causal relations, subsequent awareness necessarily involves conceptualization but without reifying either conditions or their effects, and without regarding them in terms of the grasper-grasped distinction. It is, as the texts tell us again and again, a view of the dependent as the perfected—a causal process that is, however, unbound to its unreal conditions or end products. As such, it marks a realm of agency and freedom, in the sense that the bodhisattva's perception and speech under the subsequent awareness—stemming as they do from non-conceptual awareness in which there exists no intentional object as a locus for conceptualization—align not with an allegedly given objective world (a world that "forces" itself on the mind) but with her "aspirations," understood here as reality-forming.

It should be noted that this outlook also presents an inversion of the typical view of how non-conceptual states stand in relation to the overall understanding of language. Here, non-conceptual experiences, rather than representing a complete break with language, serve, through the mediation of the subsequent awareness, as the framework against which ordinary language and perception are understood and measured for what they truly are. That is, the ineffable—insofar as it can be translated back into an exhaustive knowledge of causal relations—is regarded not as that which stands beyond the limits of expression, in splendid isolation, but as a necessary key to understanding the true function of language (I return to this below).¹⁷

I.2. The Yogācāra Notion of What Is a Concept and Contemporary Nonconceptualist Theories

The Yogācāra notion of subsequent awareness also opens up interesting possibilities for thinking about the relations between non-conceptual and conceptual content. Within the contemporary philosophical discussion of the possibility and role of non-conceptual states, proponents of nonconceptual content (henceforth non-conceptualists) find themselves, at some point, needing to account for how this content interacts and is integrated with our conceptual knowledge of the world. These accounts, however diverse and intricate, typically operate under the assumption that nonconceptual content is a primitive experiential level into which second-order conceptual content is later integrated (this is, of course, a gross generalization, and I will discuss at least one exception below).¹⁷

The Yogācāra, however, seem to operate under a different set of presuppositions about the relations that hold between non-conceptual and conceptual content. As we have seen, the Yogācāra sources seem to understand possible experiences under a tripartite model that consists of (1) ordinary conceptual experience, (2) non-conceptual experience as a distinct and independent experiential mode that by definition cannot come into contact with or function as the foundation of second-order conceptual experiences, and (3) a state in which the non-conceptual, as the absence of concepts, and conceptual content are somehow integrated under subsequent awareness (more below on how this is possible), which is, however, distinct from ordinary experience (insofar as non-conceptual experience is seen to affect and transform conceptual experience).

Another stark difference between the Yogācāra view and contemporary non-conceptualist theories has to do with their respective definitions of the scope of the non-conceptual realm and, by extension, of the conceptual realm. Like contemporary non-conceptualists, the Yogācāra wields a notion of non-conceptual content that is essentially contrastive in nature, but what it contrasts with in the case of the Yogācāra, namely vikalpa, is a notion that includes also non-propositional elements. This suggests that the Yogācāra broadens the realm of conceptual content to include not only propositional knowledge—which is the contemporary non-conceptualist's core criterion for conceptual content—but also certain non-propositional elements. This is because the Yogācāra understands vikalpa as responsible not merely for elaborate discursive thought and behavior but also for deep meaningyielding epistemic distinctions (such as the grasper-grasped dualism).

It should be noted that, in contrast to the later thought of the Buddhist epistemologists, the early Yogācāra does not require drawing any theoretical distinction between the operation of vikalpa as the cause of dualist distinction and as a more elaborate discursive activity, as these are all viewed as occurring along the continuum of the causal mental operation of the storehouse consciousness. As such, vikalpa is seen to be involved in the most fundamental levels of dualistic distinctions and concept formation that take place in the storehouse consciousness, the latter manifesting as particular "impressions of speech" (abhilāpavāsanā), which are understood to be causally efficacious. Note here that, according to the Yogācāra, conceptual linguistic activity is itself a causal phenomenon: it is causally induced and causally effective (I return to this issue below). These impressions of speech are seen both to bring about and in turn to be recursively informed by linguistic conventions, categories, and conceptual proliferations (prapañca), which are responsible not merely for overt propositional thought but for the shared elements in our cognitive experiences.¹⁸ As such, they are seen as constitutive of the objectified perception of the external world as an intersubjective realm, in a way that seems to be independent of any accompanying manifested propositional knowledge.¹⁹

Now, as it stands, this Yogācāra account deems irrelevant a great deal of the long-standing debates between conceptualists and non-conceptualists—for instance the latter's argument that the content of perception is more finegrained than the content of propositional attitudes, so that I can be aware of more shades of color than I have concepts for.²⁰ Under the Yogācāra framework, the very discrimination between different shades of color (even in the absence of any propositional content describing these differences) is already permeated by vikalpa, and therefore cannot be identified as non-conceptual.

That said, while the Yogācāra notion of vikalpa does not align neatly with what the non-conceptualist would understand as conceptual, these two frameworks are not entirely incommensurable. In the reminder of this article I would like to examine how some contemporary accounts of nonconceptual content and conceptuality can help us make sense of the Yogācāra account of these types of content, especially in the domains of linguistic performance. As I noted earlier, what we seek to explain here is above all the somewhat paradoxical view according to which, under the Yogācāra notion of subsequent awareness, propositional attitudes can be maintained without assuming that they involve vikalpa.

II. The Yogācāra Understanding of Linguistic Performances under the Subsequent Awareness

II.1. The Yogācāra and Non-conceptualist Accounts of Skillful Activity

In a paper first published almost twenty years ago²¹ and later revised,²² Adrian Cussins presented a contemporary account of non-conceptual content and an attendant theory of action that, as we will see, are highly relevant to the understanding of the Yogācāra worldview. In brief, Cussins argues that experiential content should not be understood as governed solely by notions of truth and truth conditions (with their attendant theory of semantics and an emergent notion of objectivity that he identifies as relating strictly to conceptual content),²³ but rather should be viewed primarily in terms of the agent's capacity to act upon certain circumstances (these circumstances in turn understood in terms of the possibilities they afford for action). This abilitybased understanding of experience allows him to delineate another kind of representational non-conceptual content, which is strictly success-governed, irrespective of its truth conditions as given under a theory of semantics.

For presenting these two modes of experience, Cussins turns to a rather nifty example of speeding on his motorcycle through the streets of London and being pulled over by a policeman who asks him the very philosophical question "Do you know how fast you were traveling?" and the line of thinking it provokes. The policeman's question (and the answer provided by his speedgun), Cussins points out, presupposes propositional knowledge, whose content is constitutively governed by the norm of truth. In this capacity such knowledge presents the world as a realm of reference—as conceptually consisting of particular objects, properties, relations, et cetera—with respect to which the truth of propositions may be determined. This is contrasted with a second kind of activity-based experiential knowledge: non-conceptual content that lacks any 'objectual', propositional knowledge, and is radically situationspecific and success-governed (the embodied and environmental knowledge of the motorcycle speed as function of epistemically sensitive adjustments, of which the agent is aware and for which he is held responsible).²⁴

Cussins also argues that this kind of knowledge also operates under conditions of normativity, intention-free, whose norms can be specified as the various "mundane structures of guidance-in-activity"—for instance in the case of finding one's way across an unfamiliar room:

This structure is the afforded paths or trails through the environment of the room: the activity trajectories that are afforded and which are bounded by regions of increased resistance (the edges of the trail). The pattern of trails fixes a distinction between skilled and unskilled, or competent and incompetent, activity in the environment, whatever one's intention or propositional goal may be in moving through the space.²⁵

The argument here, we can see, aligns closely with the "richness-ofexperience" argument applied by the non-conceptualist with respect to perception. Considering the teleological aspect of experience, Cussins points out that practical coping consists in more than what can be explicitly and propositionally specified about its goals. (I will return to the normativity conditions for skillful action when I deal with the inclusivist conceptualist stance below.)²⁶

On its face, Cussins' account of practical coping as involving nonconceptual content may appear to be, at bottom, a mere variant of a pragmatic theory of action (with its implicit pragmatist theory of meaning), yet what distinguishes Cussins' account from, say, Peirce's pragmatist account of action is that for Cussins objectivity is not merely bracketed but deemed irrelevant to this mode of coping. Non-conceptual content, according to Cussins, does not involve an experience or a notion of objectivity (but only possibilities of affordance for action, about which the agent might be mistaken). Objectivity emerges only once conceptual content enters the scene.

The relevance of this last point to the Yogācāra understanding of subsequent awareness (and the school's devaluation of externality) seems striking, but we should tread with caution. While Cussins' model does not appear to give logical primacy to either one of these types of content (since he understands both as action-governed and capable of standing in different temporal and causal relations to each other), his model is still far from the Yogācāra's ultimate view of these types of contents as stand-alone domains of experience.

Nonetheless, his account of non-conceptual content can help us draw out the distinctive features of the Yogācāra account of these types of content, and help make sense of how, under the subsequent awareness, propositional attitudes can be maintained without assuming that they involve vikalpa.

For the Yogācāra, as we have seen, the non-conceptual experience of a bodhisattva attests—under the subsequent awareness—to the constructed nature of objectively given phenomena, and the traditional notion of objectivity is replaced by an exhaustive knowledge of causal relations that construct and govern phenomena.

Using Cussins' terms we may say that, under these conditions, the bodhisattva is operating under an ability-based rather than a truth-governed understanding of experience, so that efficacy (underlined by knowledge of causality) rather than a notion of objectivity guides her actions. And it is worth noting that the same goes for language use: under the subsequent awareness as knowledge of causality (which applies also to how concepts come about), the bodhisattva's use of concepts manages to disengage them from their allegedly objective referential ground, instead understanding them solely in terms of their causal underpinning and their intended effects. (I return to this toward the end of this article.) Language use hence becomes a controlled causally efficacious activity (successrather than truth-governed). It is therefore the Yogācāra understanding of action solely in casual terms and the possibility of an exhaustive knowledge of such a casual framework on the one hand, and the view of language as casually efficacious on the other hand, that allows the school to understand propositional attitudes in the same terms that Cussins reserves only for the description of nonconceptual content. Conceptual activity, the Yogācāra seems to tell us, is ultimately just another mode of action. (Realizing this, however, is only possible following non-conceptual experience, which liberates language from its referential ties to an objectively given world.)

On the face of it, it is this feature (rather than the more plainly visible idiosyncratic soteriology of the Yogācāra) that makes the school's understanding of conceptuality incompatible with most contemporary nonconceptualist theories—insofar as the latter equate expression in ordinary language with conceptual articulation and contrast it with skillful "in flow" coping, which is seen to be non-conceptual.²⁷

Here, rather surprisingly, we find that Yogācāra shows greater affinity with the perspectives on conceptual articulation offered by conceptualists—but, I should note, only with conceptualists who, so to speak, lower the bar on what concepts amount to and what understanding entails, so that these notions can include within their purview skillful in-flow coping (linguistic performance not excluded). To distinguish stances of this kind—that is, those that assimilate practical coping and the realm of the conceptual—from other conceptualist stances, I will henceforth call the former "inclusivist," or "conceptual inclusivism."²⁸ And in what follows I will explore how this approach can further contribute to our understanding of the Yogācāra use of language under the subsequent awareness.

II.2. Inclusivist Conceptualism

II.2.1. Alva Noë and the inherent "fragility" of practical engagement

A variant of what I term the "inclusivist" approach is presented by Alva Noë.²⁹ Pointing out the relevancy of Noë's thought to the Buddhist notion of what a concept is, Evan Thompson (2018, p. 60) summarizes Noë's view as follows:

Following Wittgenstein, his view is that concepts are 'techniques by which we secure our contact with the world,' that understanding consists of the mastery of such techniques, and that there are many modes of understanding, of which judgement is only one. We should not make judgement be the model for all understanding. There are forms of understanding, such as the practical understanding displayed in skillful action, in which we are not able, as a general rule, to formulate judgements adequate to capture what we do. Nevertheless, Noë argues, such modes of understanding are conceptual, because they require the use of concepts, understood not as mental representations of categories, but rather as techniques for having access to the world.

Thompson further observes that Noë's understanding of concepts as enabling a kind of practical understanding displayed in skillful action may even be susceptible to the Yogācāra idea of efficacious action without a graspergrasped duality. Exploring this observation further, I want to note, first, that for the inclusivist conceptualist to be truly open to this Yogācāra picture, he must be willing to submit to the idea that concepts are "techniques" all the way down, that is, that they are essentially no different from any other mode of practical coping. This is not trivial. The litmus test for determining whether the inclusivist accepts this is, it seems to me, the way in which he understands conceptuality in the specific context of linguistic performance. Can the conceptual articulation at play in meaningful utterances—the conceptual activity par excellence—be seen as skillful and even mindless activity? And if so, what—if anything—remains of the possibility of nonconceptual content?

Noë seems to answer the first question in the affirmative, seeing conceptual use in ordinary language as primarily a form of practical understanding, a "know how," which does not necessarily require a "knowing that"—that is, an explicit propositional judgment mode of understanding. Underlying Noë's view is a strong Wittgensteinian understanding of the meaning of language as a function of its use (I return to this below). Thus, in his view, even misunderstanding—an event that we might naturally view as directing us to the normativity of language use (i.e., seeing it as a rule-governed activity)—is accounted for within language as a skillful practice (i.e., a rule-using activity) rather than requiring us to shift into a judgment mode of understanding.³⁰

This account forms part of Noë's more general view of what he calls the inherent "fragility" of practical engagement, an idea presented in response to the tendency of the non-conceptualist to identify concept-use strictly with a judgment mode of understanding, and thus to treat it as antagonistic and detrimental to the continuation of in-flow skillful action (Noë has Hubert Dreyfus in mind here).³¹ For the inclusivist, by contrast, disturbances of flow do not indicate a shift into a different mode of understanding; rather, in Noë's words, they are "business-as-usual" for skillful engagement:

To go wrong is not, as a general rule, to stop playing the game—it is not the game's abeyance—it is rather a moment in the development of play. But let's go back to language. We don't stop communicating when we fail to understand each other. At least that is not usually the case. Misunderstanding is an opportunity for more communication. Clarifying, reformulating, trying again, like criticism, are things we use language to do. The fragility is intrinsic and manifest. It doesn't mark out the game's limits. It marks one of its modalities.³²

Specifically in the case of discursive performances as skillful engagement, Noë argues, disturbances are not breaking points that mark a shift into a judgment mode of understanding (though they may trigger such a shift); instead, they merely stand for another way in which concepts serve as techniques to get a hold on our circumstances. Criticizing the inclusivist on the grounds that his account leaves no room for non-conceptual content (so as to account for skillful action) would therefore beg the question—because, for him, skillful action is fully accounted for within the conceptual realm (as a mode of understanding). Presupposing non-conceptual content is simply superfluous.³³

Bracketing the question of whether this is an adequate reply to the nonconceptualist, it should be clarified that Noë is not arguing that language use should be reduced to a practical form of understanding, but that it incorporates this mode of understanding alongside other modes of understanding, including, among others, the judgment mode,³⁴ and, moreover, that disturbances of discursive engagement—its so-called fragility—do not indicate the exclusivity of one mode over the other. But regarding this latter claim, one might object that disturbances of discursive practical engagement, such as miscommunication, insofar as they are an exception to the norm (because we understand each other more often than not), presuppose normative knowledge of language use: this knowledge is necessary for us to acknowledge an error. And simply to appeal to the Wittgensteinian notion of language use as rule-using rather than rule-governed would be too general a response for this purpose (to do justice with Noë it should be said that the concern I raise here is peripheral to the main argument of his paper). To complete our understanding of concepts as techniques "all the way down" we therefore still need to figure out how normativity—conceptual normativity in particular—figures in discursive performance as skillful practical engagement.

II.2.2. Joseph Rouse on the conceptual normativity involved in language use as practical coping

A step in the direction of such an account is proposed by Joseph Rouse,³⁵ who argues for the possibility of accounting for conceptual articulation and understanding as itself a form of practical-perceptual coping, with specific attention to the conceptual normativity that is at play. Rouse's claim, it should be clarified, is not that discursive performances are somehow interrelated with the practical-perceptual responsiveness to circumstances, but that they are themselves forms of practical-perceptual coping, much like any other kind of skilled coping.

Rouse presents his account through a reframing of the DreyfusMcDowell debate about whether practical-perceptual coping skills are conceptually articulated, and their related disagreement over what is at stake in distinguishing conceptual understanding from non-conceptual capacities.³⁶ Rouse's main observation is that their positions are orthogonal rather than contrastive, and he proceeds to offer his own synthesis, which merits a detailed examination.

One easy way into Rouse's intricate argument proceeds through his treatment of Dreyfus' famous example of the blitz chess played at a speed of two seconds per move, the point of which is to illustrate the absence of conceptual understanding from an experts' mindless coping with her environment (which is therefore characterized as non-conceptual). The example and the argument it supports, Rouse points out, pose a challenge only to descriptive accounts of conceptual articulation (i.e., accounts that take conceptual content to be something actually present or operative in specific performances by concept users). Yet John McDowell's concern with conceptual articulation, Rouse points out, is essentially normative.³⁷ The issue at stake, therefore, is that the blitz chess game (or, for that matter, any chess game) must rely on conceptual normativity—for instance on the various concepts relevant to the game, such as pieces, moves, et cetera—and this is so regardless of whether the player has these concepts in mind, in the sense of explicit representation, while playing. As Rouse indicates, what matters for a normative account of conceptuality is only if one's performance is or can be held accountable to the relevant standards in the right way.³⁸

Having framed the two sides of the debate in this way, Rouse proceeds to offer his synthesis. His final focus is on conceptual understanding as manifested particularly in discursive performances. Here he builds on Dreyfus' notion of skillful coping, but without assuming that such coping is necessarily preconceptual or non-conceptual; and while he embraces McDowell's general normative approach to the conceptual domain, unlike McDowell he does not see it as self-evidently pervasive also in perception.³⁹ Rouse therefore proposes to account for conceptual understanding as itself a form of practical-perceptual coping (much like other kinds of skilled coping) within which, however, conceptual normativity is also at play.⁴⁰ This kind of practical-perceptual coping, he argues, is present even in the very basic ability to utter grammatical and meaningful expressions in ordinary language, a practice that does not always require an explicit presence of mind (for instance in conversational situations where we 'discover' what we are saying as we say it), yet is not thereby rendered non-conceptual.⁴¹

This coping dimension of language use, Rouse argues, is more fundamental than philosophers typically care to acknowledge, and in this respect he is critical of what he sees as a "very thin" philosophical conception of language, one that is not attuned to the "practical-perceptual skills of speakers and listeners, their bodily involvement in the world, and the social-institutional settings in which their skills are exercised."⁴² Such aspects of linguistic performance as coping are manifest, for instance, in the way in which discursive responsiveness is embedded within complex social relations,⁴³ and in the important practical-perceptual skills and bodily involvement that take part in language acquisition, particularly first-language acquisition.⁴⁴ Framed in this way, language is re-configured as a practical-perceptual way of coping with a discursively articulated "environmental niche" (rather than as a formal structure that merely interfaces with practical-perceptual coping).⁴⁵

Mitigating what might seem like a highly reductive view of language as a form of biologically adaptive behavior is Rouse's insistence on including conceptual normativity as an integral part of this account. His concern here is how this level of description may be integrated with the conceptual contentfulness of discursive practice, in a way that makes sense of the gap between what is meant and what is sometimes actually encountered in coping (a gap revealed, for instance, by the explicit acknowledgment of an error by a skilled practitioner, and in formal analysis by the sense/reference distinction).⁴⁶ Such a gap is only affordable, Rouse notes, once we accept the normativity involved in conceptual understanding.

Rouse therefore attempts to show how this normativity can be at play not merely with respect to or following practical-perceptual coping (as if they were two separate domains, as the non-conceptualist would hold), but as intrinsic to it, taking the practical-perceptual coping with our surroundings to be "itself the locus of conceptual articulation."⁴⁷ Rouse's unpacking of this view begins by acknowledging that conceptual articulation (and the attendant conceptual normativity at play) is what allows discursive performances to be decoupled from their immediate nonverbal circumstances (so that a gap can open up between what we mean and do)—but only insofar as these performances are mediated by intra-linguistic connections and are accountable to other intra-linguistic performances.⁴⁸ For Rouse, then, language is itself a mode of coping, but one that must possess a semiautonomous status (and the qualifier "semi" makes all the difference here) afforded by conceptually articulated normativity.

To give a few examples, this semiautonomous capability of language—as practical-perceptual coping—is expressed in the vocative character of discourse.⁴⁹ While on the one hand this feature emphasizes the complex practical skills involved in language use, on the other hand "the vocative aspect of utterances also makes their links to past and future utterances internal to what the utterance itself does. These links reinforce the partial decoupling from surrounding circumstances that makes discursive practice semiautonomous."⁵⁰ Another indication of the conceptually articulated normativity at play within coping is the rather open-ended teleology of perceptual-practical discursive activity—in other words, our ability to do more, to mean more, discursively speaking, than a mere reaction to circumstances (in contrast, for instance, to non-linguistic forms of life). Such an ability, Rouse seems to say, should not be seen as a disengagement of language from practical-perceptual coping, but rather as a broadening of what such coping entails.⁵¹

To re-cap: for Rouse, discursive performances themselves are instances of practical-perceptual coping, and thus can be seen as "in flow" mindless activity (which does not necessitate a conceptual/non-conceptual divide). As such, however, they are pervaded by conceptually articulated normativity, which, far from being an obstacle to their functioning, in fact enables their unique mode of operation (manifest, for instance, in decoupling from mere responses to circumstances, relatively open-ended teleology, interrelations with other linguistic performances, etc.).

II.3 Inclusivist Conceptualism versus the Yogācāra Understanding of Language Use

Returning to Yogācāra, Rouse's account of discursive performance as coping illuminates some of the features of the Yogācāra view of language use—under the subsequent awareness—as ultimately just another mode of action. In particular, it helps us think anew about some of our difficulties in making sense of this view of language as both characterized by the use of concepts and, at the same time, entirely free of vikalpa (not merely in the sense of a grasper-grasped distinction, but in the broader sense as employing any conceptual distinction). In light of Rouse's account, such difficulties appear to arise only when we approach the Yogācāra view through a framework that presupposes a demarcation between non-conceptual and conceptual content, wherein perception and in-flow coping fall neatly on one side of the line and language use on the other. The inclusivist conceptualist critique of this demarcation assists us in realizing that what this kind of framework takes to be concepts and what it understands as non-conceptual is not at all what Yogācāra had in mind. In fact we might be talking at cross purposes when we continue to apply these terms in this way to the Buddhist materials.

Another ramification of the inclusivist conceptualist view is that by bringing language under practical-perceptual coping it stands to show us that the Yogācāra account of language is perhaps more accessible to us than we imagine, a matter not merely of admitting the school's soteriology but, more fundamentally, of getting our theory of ordinary language right.

So is the Yogācāra account of language a variant of conceptual inclusivism? Well, it is closer to this view than to non-conceptualism (insofar as the latter takes non-conceptual skillful action as antagonistic to language use). But ultimately the answer is no. The main difference between the Yogācāra stance and conceptual inclusivism stems from the Yogācāra soteriology and is expressed in the radical degree to which the school regards discursive performance, and language use in general, as capable of determining the very circumstances to which language responds. I will clarify.

The understanding that emerges from Rouse's account and is shared by Noë about what a concept is, amounts eventually to the Wittgensteinian idea of a concept as primarily a kind of a tool, a technique, and of the meaning of language as a function of its use (although neither Rouse nor Noë is concerned here with the question of meaning). Under this framework, language use is seen as deeply involved in the process of determining and negotiating what is the case. But while this account turns away from a representational view of language, it still assumes a certain independently given state of affairs to which language is responsive (indeed, this independent state underlies the view of language as "coping").

Consider, for instance, Wittgenstein's famous example, in his Philosophical Investigations, of the language of builders and the notion of a language game (I, 2, 7–8, 19–20, etc.), or, better yet, consider the opening lines of Tom Stoppard's Dogg's Hamlet, a play that draws on Wittgenstein's example to create a work whose performance teaches the audience the language in which it is written:

Translation from 'Dogg' language into English is given in square brackets where this seems necessary.⁵²

(Empty stage)

baker:

(Off-stage) Brick! [*Here!]

(A football is thrown from off-stage left to off-stage right.)

baker:

(Receiving ball) Cube. [*Thanks.]

(abel, off-stage, throws satchel to stage left. abel enters. He is a schoolboy wearing grey flannel shorts, blazer, school cap, etc., and carrying a satchel. He drops satchel centre stage and collects the other which he places with his own. abel exits stage right and returns with microphone and stand which he places down stage. The microphone has a switch.)

abel:

(Into the microphone) Breakfast, breakfast… sun—dock—trog.… [*Testing, testing… one—two—three]

(He realizes the microphone is dead. He tries the switch a couple of times and then speaks again into the microphone.)

abel:

Sun—dock—trog—pan—slack… [*One—two—three—four—five]

(The microphone is still dead. abel calls to someone off-stage.)

Haddock priest! [*The mike is dead!]

(Pause, baker enters from the same direction. He is also a schoolboy similarly dressed.)

baker:

Eh? [*Eh?]

abel:

Haddock priest.

baker:

Haddock?

abel:

Priest.

(baker goes to the microphone, drops satchel centre on his way.)

baker:

Sun—dock—trog—

(The mike is dead, baker swears.)

baker:

Bicycles!⁵³

Like a game whose rules are learned through playing it, language learning and language use—both essentially public activity—also hinge on active interaction with its users. But while the conceptual normativity at play here is regarded as emergent, determined by use and to some extent contingent content-wise, the use of language is nonetheless still primarily responsive—it evolves around and reacts to the event of an object thrown from off-stage, or a malfunctioning piece of equipment—a given that impinges itself on our perception.

Here is where the Yogācāra understanding of language use under the subsequent awareness distinctly parts ways with the inclusivist conceptualist account. As we saw above, under the subsequent awareness the bodhisattva's language use is seen to be free of any objectively given referential framework.⁵⁴ To put it more accurately, for the skilled bodhisattva the final measure of language is its efficacy, but as such its alleged referentiality does not determine its application. Rather it is the other way around: its referentiality is emergent, a by-product, so to speak, of efficacious language use. Language here is not merely responsive to or involved in but entirely responsible for creating—in the most radical sense—the circumstances of its own involvement. While this account is still not incompatible with the inclusivist conceptualist account, its underlying metaphysics would make it a bitter pill to swallow for the inclusivist, as it requires a revision of what skillful coping is essentially about—namely, not merely a responsivness to and a negotiating with, but a reality-creating activity.

Furthermore, and here we come full circle to our starting point, the Yogācāra view of language use as a reality-creating skillful activity in turn reflects back on the school's understanding of what non-conceptuality (nirvikalpa) entails: it is not the non-conceptualist idea of a content or state on top of which or alongside which conceptual activity may take place, but the complete absence of any linguistic and conceptual activity, and nothing more.