[Page 137]

WHAT ARE THE GROUNDS OF EXPLICATION?:
A BASIC PROBLEM IN LINGUISTIC ANALYSIS AND IN PHENOMENOLOGY

In this paper I will attempt to discuss linguistic analysis and phenomenology accurately so that the adherents of each can agree with what I say, and yet also the discussion of each method must be understandable to the adherents of the other. If I can really do that, the basic similarities will appear. I will attempt to state some propositions that apply to both frames of reference. The similarities which these propositions state are basic aspects of philosophic method, and they also pose a major problem.

The problem, as I see it, concerns the grounds of explication. In both methods the main assertions are founded neither on formal logic nor on observed relationships. They are based on an 'implicit' knowledge (if we call that 'knowledge'): on what is 'implicit' in experiencing, living, and acting in situations. How are philosophic statements founded on something 'implicit'? Such statements are called 'explications'. What is the basis for asserting and evaluating them? What criteria are possible for such statements?

The problem, as I will try to show, does not lead backward to a reassumption of 'external' criteria or constructs, but rather opens a new area of philosophic study.

I will not generalize away the differences between various linguistic analysts and between different phenomenological philosophers. (In fact, these differences bring home the problem of criteria.) I will limit myself to specific formulations from one philosopher in each mode. Generalizations of a lowest common denominator are not really usable without their detail. On the other hand, even one bit of detailed discussion, if closely examined, displays the method. I will use a few excerpts from the writings of Austin, and then from those of Sartre.

Propositions applicable to both methods will be formulated [Page 138] first as they arise from my excerpts of linguistic analysis. In this way the propositions will be connected to their detailed employment there, and yet phenomenologists will recognize them. Later, in my discussion of Sartre, I refer back to them. Thereby I may be able to show how the methods are similar and pose a similar problem of explication.

If indeed the methods are similar in the ways I try to show, then my later discussion of explication can be considered an instance of either method and should carry both methods forward into a new and central problem area. I realize that this is a considerable program, but if the methods are specifically similar in this respect, then the program is possible.

I

I turn first to that kind of philosophizing which proceeds from ordinary language. Austin approaches a question of moral philosophy as follows:

. . . a study of . . . 'excuses' . . . will contribute in special ways . . . to moral philosophy in particular . . . (p. 125) [1]
When . . . do we 'excuse' conduct . . . In general, the situation is one where someone is accused of having done something . . . Thereupon he, or someone on his behalf, will try to defend his conduct or to get him out of it.
One way of going about this is to admit flatly that he, X, did do that very thing, A, but to argue that it was a good thing . . . To take this line is to justify the action . . .
A different way of going about it is . . . to argue that it is not quite fair or correct to say baldly 'X did A' . . . perhaps he was under somebody's influence, or was nudged . . . it may have been partly accidental, or an unintentional slip.
. . . briefly . . . we admit that it was bad but don't accept full, or even any responsibility (PP, pp. 123-24).

In this excerpt Austin makes some discriminations: Defense by justification of an action is distinguished from defense by [Page 139] disclaiming responsibility. Then different kinds of disclaimers of responsibility are distinguished.

The excerpt shows that the question of responsibility (think of the traditional question of 'free will') is here discussed in a context. Linguistic analysis operates in the context of specific situations (" . . . perhaps he was under somebody's influence, or was nudged . . . ").

Proposition 1: Philosophic terms are examined and used in the context of living and acting in the world. The use of each term marks a discrimination in that context.

What one cannot do (is advised not to do) in this mode of philosophy is to work with abstractions purely theoretically.

For example, 'responsibility' cannot be discussed in the manner of the 'free will problem' of some traditional philosophies: 'voluntarily' and 'involuntarily' are not 'contraries' such that one of them always applies to any action.

. . . given any adverb of excuse, such as 'unwittingly' or 'spontaneously' or ' impulsively', it will not be found that it makes good sense to attach it to any and every verb of 'action' in any and every context:
For example, 'voluntarily' and 'involuntarily': we may join the army or make a gift voluntarily, we may hiccough or make a small gesture involuntarily . . .
'Voluntarily' and 'involuntarily', then, are not opposed in the obvious sort of way . . . The 'opposite', or rather 'opposites', of 'voluntarily' might be 'under constraint' of some sort, duress or obligation or influence: the opposite of 'involuntarily' might be 'deliberately' or 'on purpose' . . . (PP, pp. 138-139).

If you ask about a phrase in this paper whether I wrote it 'voluntarily' you imply something rather unusual about my professional circumstances or about my editor. But, you don't imply the contrary (you imply something totally different) if you ask whether I wrote that phrase 'involuntarily' (perhaps it slipped out and went unnoticed).

If we examine each term as tied to the circumstances its use discriminates, then we are not examining the terms as constructs. As constructs these two terms above would be contraries. As used, [Page 140] they discriminate different and not directly related aspects possible in living.

But what of the vast main body of actions which involve neither of these special aspects? Are they done 'voluntarily' or not? Is one 'responsible' in general for actions (as per 'free will') or not? Obviously, the question is nonsense if one anchors the terms 'voluntarily' and 'involuntarily' each to their special contexts.

Does this mean that the famous and rich problem of human freedom has been avoided by a sleight of hand? Not at all. The problem has become many more specific problems, anchored to discriminations in the world in which we live and act. If we ground each term in the aspects it marks we can also discuss:

. . . the detail of the complicated internal machinery we use in 'acting'—the receipt of intelligence, the appreciation of the situation, the invocation of principles, the planning, the control of execution and the rest (PP, p. 127).

If we proceed in this way, nothing need be lost.

But, using such distinctions, can one conclude anything on a question, for example, a question of moral responsibility? Austin offers an example: Finney, an attendant in a mental hospital, ran scalding hot water into a bathtub while a patient was in the tub, killing the patient. Was Finney 'responsible'? Austin quotes his attorney:

. . . if the prisoner, knowing that the man was in the bath, had . . . turned on the hot instead of the cold water, I should have said there was gross negligence; for he ought to have looked to see. But . . . he had told the deceased to get out, and thought he had got out. If you think that indicates gross carelessness, then you should find the prisoner guilty of manslaughter. But if you think it inadvertence not amounting to culpability—i.e., what is properly termed an accident—then the prisoner is not liable (PP. p. 145).

For Austin the excerpt illustrates a "very free use of a large number of terms of excuse . . . several as though they were . . . equivalent when they are not," thus showing that 'ordinary use' in people we observe often needs sharper discriminating. But the excerpt also shows that discussion and conclusions are based on the discriminations marked by our uses of terms.

One could now argue that 'free will' is 'assumed' in this [Page 141] excerpt, since it concerns the fact that Finney didn't 'choose' the hot water tap knowing the patient was in the tub; we seem not to question here—but rather to assume—that, in general, a man can choose. Is there a general question of freedom 'above and beyond' (or 'underneath') the specific discriminations of various kinds of ordinary acting and responsibility?

By saying or asking about 'voluntarily' we imply aspects of coercion in the situation within the ordinary human world ( not some 'underlying' constant coercion of 'determinism'). By saying 'deliberately', we discriminate something noticeable in an action (not some 'underlying' construct of 'freedom').

Proposition 2: There are no entities, constructs or determinants assumed to be 'behind', 'beneath', or 'over and above' the world in which we live and act. There are no 'external' principles or criteria.

But if there are no more basic, underlying principles against which to evaluate language, must linguistic analysis simply accept the 'prisons of the grammarians', the assumptions in how language happens to slice and render the world? Not at all.

To examine and evaluate assumptions, constructs, 'models', is a main task of linguistic analysis. But this critique does not invoke superordinate principles. Rather, the critique leads models back to (and limits their use to) the specific circumstances which their use can mark. For example, take the assumption (the 'model') of 'free will' and of 'all actions'.

. . . 'doing an action', as used in philosophy, is a highly abstract expression—it is a stand-in used in place of any . . . verb with a personal subject, in the same sort of way that 'thing' is a stand-in for any . . . noun substantive, and 'quality' a stand-in for the adjective . . . notoriously it is possible to arrive at . . . an over-simplified meta-physics from the obsession with 'things' and their 'qualities'. In a similar way . . . we fall for the myth of the verb (PP, p. 126)
. . . We take some very simple action, like shoving a stone, . . . and use this . . . as our model in terms of which to talk about other actions and events . . . even when these other other actions are . . . much more interesting . . . than the acts originally used in constructing the model . . . (PP, p. 150).
. . . to say we acted 'freely' . . . is to say only that we acted not unfreely, [Page 142] in one or another of the many heterogeneous ways of so acting (under duress, or what not) . . . In examining all the ways in which each action may not be 'free', i.e., the cases in which it will not do to say simply 'X did A', we may hope to dispose of the problem of Freedom (PP, p. 128).

Proposition 3: Arguments are not conducted or evaluated critically on general theoretical grounds. Instead, assumptions and models are critically evaluated in terms of what they discriminate in living and acting in the world.

An overextended use of an expression is called 'misleading' even when perfectly sound logical implications were drawn from it by logical necessity. To formulate what the use of an expression discriminates, we are not led by its logical implications, but rather by the 'implications' of its use. Not the expression as such, but its use, 'implies' the discriminated aspects in situations.

Let 'p' stand for an expression. We say 'p' only when facets A, B, C obtain. Therefore, when we say 'p', we imply facets A, B, C.

Austin writes: "not p . . . but asserting p implies . . . By asserting p I give it to be understood that . . ." (PP. p. 32).

Thus the main assertions of linguistic analysis concern an 'implication' by an activity (using words).

Proposition 4: The main philosophic assertions state 'implications' of a different sort than logical implications. They state 'implications' of an activity.

And now the question of criteria: When (as in most of its central assertions) linguistic analysis states what aspects of the world the use of a word marks, i.e., what the use 'implies', how may we tell when such statements have correctly (or adequately, or well) formulated these circumstantial aspects?

The question is illustrated by a celebrated [2] disagreement between Austin and Ryle in a case of this sort. Ryle said:

In their most ordinary employment 'voluntary' and 'involuntary' are used . . . as adjectives applying to actions which ought not to [Page 143] be done. We discuss whether someone's action was voluntary or not only when the action seems to have been his fault. . .

There is a contradiction in the fact that to "join the army or make a gift" (Austin's examples of when we use 'voluntarily') are not actions "which ought not to be done."

What criteria are there for statements that explicate such 'implications' as these?

Since not the word, but its use, does the 'implying', our question of criteria depends on what sort of 'implication' an activity like using can have. As we already said, it differs from logical 'implication':

. . . 'implies' must be given a special sense . . . (PP, p. 32).

It is not an 'analytic' implication. If it were analytic, then whenever x is y,

y must be either a part of x or not any part of it . . . (as) would be the merest common sense if 'meanings' were things in some ordinary sense which contained parts in some ordinary sense (PP, p. 30).

The 'model' of 'analytic' implication is criticized by leading it back to just those situations it discriminates when used properly, situations where there are 'parts' which can be 'included'. Austin calls it a "shabby working model" which "fails to fit the facts that we really wish to talk about" (PP, p. 30). His italics in this sentence indicate that the circumstances of using a word do not have given 'parts'.

Proposition 5: The activities whose 'implications' the philosophy states cannot be assumed to consist of parts or units in some ordinary sense.

But neither are these 'synthetic' statements. They are not based on a survey of how most people use a word, nor would such a survey be pertinent. Most people (like Finney's counsel) use language sloppily, or not as sharply as it can be used.

Let us treat statements of 'implications' of use like any other expressions. We want to examine the use of such statements by philosophers, so let us see what they do, and when.

What do linguistic analysts do with such assertions? They formulate 'rules' for when we should use expressions. Who is entitled [Page 144] to formulate such rules? The answer is: the native speaker of a language. By 'native' is marked the sort of learning which doesn't occur from a book but rather, learning in a context of living. We have learned language (not as labels for rigid objects as might be pictured in a book, each with word below), but through living in contexts in which words are used. Thus, explication statements occur when 'native' speakers spell out specific aspects of the contexts in which they have learned to use words. Such statements spell out a learning (or 'knowing') which is not already cut up into the sort of 'parts', or units or variables we use in 'spelling out'. The learning and 'knowing' of language by a native speaker (his learning or 'knowing' the circumstances) is 'implicit': Austin calls this a learning of "semantic conventions (implicit, of course), about the way we use words in situations" (PP, p. 32).

Thus the use of the word 'implicit' in this context marks several aspects:

Firstly, it marks a relation between an activity (like using) and a statement, such that what is called 'implicit' is 'in' the activity but not in the form of the spelled out units of statement.

Proposition 6: The statements are 'implicit' in the activity, i.e., not in the sense that its verbal units, 'unit-meanings' or 'representations' are 'in' it.

Secondly, 'implicit' marks a relation between activity and world again in a sense other than as concrete or represented things 'in' it.

Proposition 7: Aspects of the world (contexts) are 'implicit' in the activity, i.e., not in the sense that things (or images of them) are 'in' the activity.

Thirdly, our knowing when to use a word (our having learned to use it) is said to be 'implicit' in the activity of using the word (and in our habit or capacity to do so). This is the 'native' knowledge which entitles us to make assertions about 'when we should say . . .'. If we did not call it 'implicit', then the 'knowledge' would be the sort we have after rules are formulated. But,we do not learn language by repeating explicit 'rules' to ourselves.

[Page 145]

Proposition 8: We know 'implicitly' what the main philosophic statements formulate, i.e., not in the sense of a knowledge separable from activity (or the capacity for the activity).

We now see that explication statements are neither 'analytic' nor 'synthetic' for the same reason; 'analytic' is used when there are already defined and separable units or parts (as in formal logic), and similarly, 'synthetic' is used when there are already defined observation units (as in observing two 'associated' traits). But we do not learn, know, or use language by knowing separate, defined, unitized 'variables' of circumstances.

The Linguistic Analyst has learned what the use of a word might mark, but not as separate 'synthetically' associated variables. He proposes exemplary sentences to himself in imagined contexts, and thereby he notices newly specific [3] variables standing out, which may convince him that the example is right or impossible.

Proposition 9: Observation and experience cannot be assumed to be constituted of already given units, 'variables'.

Austin emphasizes that explicating the circumstances of 'use' is not merely a matter of observing already given variables. 'Situations', 'circumstances', 'actions', the contexts in which we 'use' language, are not given in already definite units (these would be reified verbal units). Austin says situations are capable of being "split up" along "various lines." For example, there is no hard-and-fast way of knowing what is 'an' action: ". . . what, indeed, are the rules for the use of 'the' action,an' action, a 'part' or 'phase' of an action, and the like?" (PP. p. 127). Austin indicates here that this problem must be dealt with by the same method as any other ("what are the rules for the use of 'the' . . . or 'an' . . .action"?). The criteria for unit actions would have to be found in the circumstances in which we use 'the' or 'a' or 'an action'. [Page 146] But circumstances of using do not come in already defined unit variables. Thus, to explicate the 'rules' (i.e., circumstances of use) for units of action involves the same difficulty. Austin points this out and urges us not to take as our model-cases only those simple and relatively dull circumstances in which variables are already definite units, plainly and easily given:

. . . is to think something, or to say something, or to try to do something, to do an action? . . . All 'actions' are . . . equal, composing a quarrel with striking a match, winning a war with sneezing . . . (PP. p. 127).
. . . what is an or one or the action? For we can generally split up what might be named as one action in several distinct ways, into different stretches or phases or stages . . . we can dismantle the machinery of the act, and describe (and excuse) separately the intelligence, the appreciation, the planning, the decision, the execution and so forth (PP. p. 149).
. . . it is in principle always open to us, along various lines, to describe or refer to 'what I did' in so many different ways (PP. p. 148).

The method seems to offer no criteria to decide among those various 'lines' along which to split up.

Surely, we won't let it remain at this! Can it be that these philosophers uncritically use just whatever assumptions happen to creep into the 'lines along which' they 'split up' the circumstances of use and action?

Proposition 10: The philosophical method has the problem that explications can be formulated along various lines. The explications depend partly on the given philosopher's mode of formulating, for which no justification is offered.

Philosophic questions, after all, are just those which deal with the problems of the variety of modes of conceptualizing. Are we to say that ordinary language philosophy cannot deal with just this most properly philosophic type of question?

The problem is rounded out since 'use' itself is probably an 'action' (it depends on how we split it up!). Austin's treatment of action here is therefore an inquiry into the fundamentals of how one explicates use. Austin formulates some parameters along which use or action can be split up in a variety of ways: "stretch," "phase," "stage." But he does not intend to make this into a scheme similar [Page 147] to the many schemes philosophy already has.

Proposition 11: No one scheme can be given to analyze the various parameters along which possible explications may differ. Such a scheme would only again follow models or constructs. The activities in their contexts are always more basic than any system of constructs or parameters.

Austin wants us to notice that the context is not already "split up" into given, cut-and-dried variables.

When we examine what we should say when, what words we should use in what situations, we are looking again not merely at words . . . but also at the realities we use the word to talk about . . . (PP, p. 130).
We need therefore to prise them (words) off the world, to hold them apart from and against it, so that we can realize their inadequacies . . . and can relook at the world without blinkers (PP, p. 130).

The world at which we 'relook' is thus not assumed to be already structured out for us in some hypothetical way, or as words have seemed to cut it. In the direct examination of the circumstances when we use words, we "relook" without being bound by the prisons, cooky forms, models of already given variables. Austin says we get "a sharpened perception" of phenomena "directly." The analysis of words' uses leads beyond the ways words had structured the world. We know the world 'directly' and we say of this knowing that it is "implicit." Words have no "handy appendages," "meanings" or "denotations" corresponding to them (PP, p. 29).

Proposition 12: Since activities in context are more basic than constructs or pictures, the view of meaning and cognition differs from the traditional. No longer are contents, representations, denotations, pictures, objects as referents, considered basic. The representational view of meaning is overthrown. Neither world nor activities are assumed to be given already cut up into 'handy denotations' or set given 'meanings'.

One can not argue that we simply note what is 'similar' in the different circumstances in which we use a word. Austin opposes the assumption that there is somewhere a denotion or 'respect' in [Page 148] which situations are 'similar' (PP, p. 38). "The different meanings of the word 'head' will be related to each other in all sorts of different ways at once" (PP, p. 43). And this is an argument not merely against real universals. It is an argument to show that the world of circumstances (in terms of which 'use' is explicated) is not already split up into handily packaged variables or denotations.

A striking example is the case of 'pleasure': pleasures . . . differ precisely in the way in which they are pleasant. No greater mistake could be made than . . . of thinking that pleasure is always a single similar feeling, somehow isolable from the various activities which 'give rise' to it (PP, p. 41).

There is no 'similarity' given such that we might simply look to see what is 'similar' when we use a certain word. For example, we must look 'directly' to see the circumstances in which pleasure is termed one, and those in which it is termed by various subtle specific words. Thus, while our formulations of circumstances are not arbitrary (they are systematically related to our 'looking directly') neither are these formulations governed by given 'similarities' or given variables of observation. The lesson is that the world cannot be assumed to consist of ready and waiting variables. Is Austin perhaps saying that our words and concepts do not necessarily conform to some unknowable nature of "things in themselves"? Just the opposite: he is saying that we may 'relook directly' and that we know 'implicitly'.

Proposition 13: The method involves a direct access to an experienced world not yet split into word-like or thing-like units or traits. The method systematically relates formulations of philosophy to this direct access.

Is not this an attempt to leap out of the ancient problem of the variety of ways of construing anything? If we can 'relook directly' at the circumstances words mark, and if thereby we think ourselves freed of presuppositions, assumptions, constructs, variables already isolated, etc.—if we insist that circumstances are not given with 'intelligible essences' ('similarities') of their own,—are we not at the mercy of whatever assumptions and selections creep into our formulations of this direct, 'implicit' context?

[Page 149]

How then can we evaluate a formulation (or worse: two differing ones)?

There is a seeming 'vicious circle' in evaluating all terms and propositions by what circumstances their uses implies, and then basing statements of use-implications on what? On direct looking. But we assert that the world looked-at 'directly', 'without blinkers' does not come marked out by heavy black lines into 'variables', kinds' of circumstance, 'similarities'. We assert thereby that direct looking cannot provide a foundation for the assumptions and varieties of formulations.

Proposition 14: The ultimate statements of the philosophy are said to be 'direct descriptions', formulating what is already 'implicit' in certain human activities in the situational world. Yet, it is expressly denied that what we 'look at directly' has the structure various formulations employ. Thus one both asserts and denies that the structure and assumptions of descriptions inhere in the directly looked at.

But if no ultimate structure is 'given' then why should we need for it to be? There must be a positive view of this seeming 'vicious circle'. We must accept that explication is not 'based on' something 'implicit', through a correspondence of structure and must examine what the relationship is which is here called: 'based on'.

To do so we must examine what we do when we explicate.

II

Phenomenologists will have recognized that each of my 'propositions' is basic also to their method. But I have not yet presented phenomenology so that linguistic analysts could appreciate how these propositions apply in it. Also, I have not yet shown that the problem of explication statements is central also for phenomenology.

Linguistic analysts wish to limit discussions to precise formulations. Since the method (as, I hope, I have shown) centrally involves something 'implicit', i.e., something not formulated, they find it difficult to describe their method. Phenomenologists choose the opposite order: they begin by pointing to something unthe-[Page 150]matized, pre-objective, pre-conceptual, experienced and lived but not explicitly known.

This difference in where the two methods begin leads to characteristically different common misunderstandings of each: linguistic analysis can seem 'trivial', concerned only with extant words and linguistic conventions. Phenomenology can seem 'fuzzy', concerned only with unspeakable unknowables.

These characteristic misunderstandings point up that these methods involve a relationship between formulation and the not yet formulated. One mode is misunderstood as using only what is already formulated, the other as using only the inveterately unformulable.

It is well, therefore, to say of phenomenology first that it formulates, explicates, 'lifts out', renders in structures and discriminating description, aspects 'implicitly known' to us, but not known explicitly till formulated.

Phenomenologists too, like to claim that their formulations are 'direct descriptions'. They like to say that they are not 'explaining' (rendering in construct systems) but rather, 'only' describing. Heidegger considered his formulations as explicating the basic structure 'implicit' in living and acting, yet Husserl found them incorrect. Obviously not every aspect of these differing 'descriptions' is directly founded on phenomena directly viewed by all. (See Proposition 14.)

Phenomenologists (like linguistic analysts) do not argue from theoretical models. All terms, propositions and arguments are evaluated not by their theoretical structure but by what they 'lift out' directly for us. If, as a result of the 'description' we now notice something directly, the description has done its proper work.

Phenomenology depends upon our having, in addition to terms and constructs, something directly accessible ('phenomena') not assumed to be determined or patterned by assumed constructs. The method systematically relates philosophy to this directly accessible (Proposition 13).

What is the nature of directly accessible experience or phenomena? Husserl examined only 'experience' but experiences are always 'of' something. On phenomenological grounds Husserl rejected 'psychic entities'—he just never found them. They were never directly noticeable. They are mere constructs of a certain [Page 151] kind of psychology. Experiences are always of what is seen, noticed, heard, aimed at, wanted, expected, desired, etc. Husserl found more and more that the whole human world is involved when one examines 'only' experience.

It takes artificial constructs to say that our experiences are 'in' us, that they are 'subjective', that there is 'another' world out there, in addition to 'percepts' in us. Phenomenology is thus insistently not 'subjective'. The ordinary human world is 'implicit' in experience. Heidegger and Sartre begin where Husserl gradually arrived. Experience ('consciousness') is 'being in the world'.

Sartre begins where our discussion of linguistic analysis ended, namely, the refusal to assume that our activities (action, use, thinking) or the world are constituted of thing-like entities, meaning-constructs, or representations.

(It is an error to) . . . make the psychic event a thing and to qualify it with 'conscious' just as I can qualify this blotter with 'red' (liv) [4]
We must renounce those neutral 'givens' which, according to the system of reference chosen, find their place either 'in the world' or 'in the psyche' (BN, p. li).
Consciousness (of) pleasure is constitutive of the pleasure as the very mode of its existence . . . and not as a form which is imposed by a blow upon a hedonistic material (BN, p. liv).

The point—even the choice of an example—is similar to Austin's, as already cited. There are not already given meanings, units or forms like 'pleasure'. Pleasure "is not a representation, it is a concrete event" (BN, p. liv). (Proposition 12.)

Sartre's emphasis is on the ongoing process, the activity. Representations ('objects of knowledge', 'objects of reflection') occur only as 'supported by' the concretely ongoing process. There can be consciousness of representations, but not representations of consciousness. All structures, concepts, representations, schemes and laws are to be viewed as already involving a concretely ongoing activity, and thus they can never explain or picture it.

[Page 152]

It is futile to invoke pretended laws of consciousness . . . a law is (an) object of knowledge; there can be consciousness of a law, not a law of consciousness (BN, p. lv).

Thus the concrete activity remains always more basic than any representations or laws (had only through it) and hence no ultimate laws of it can be given (Proposition 11).

Sartre's constructs ('for itself and 'in itself', and various specific pairs of terms similar to this pair) stem from Hegel. They cannot be attributed to the very nature of consciousness or the world (except insofar as with these terms . . . and with other terms, too . . . we can 'lift out' and bring to notice aspects of living which we then can see directly, i.e., even without these terms).

You may therefore view Sartre somewhat as though he were a linguistic analyst who happened to like dialectical instead of British Empiricist types of terminology. Just as linguistic analysts are not party to British Empiricist assumptions although they often employ that philosophical tradition's terms, so Sartre is not fashioning an abstract dialectic although he uses dialectical terms.

But, is there any phenomenological foundation for this choice of terms? If there were, would we not have to assume that there is something inherently dialectical (or, if we oppose it positively, something inherently non-dialectical) in the phenomena and activities themselves? As we saw, no dialectical or other such "laws of consciousness" are possible. The concretely ongoing activities are always already involved even in the holding of such laws.

Thus there is no basis for the formulative assumptions and constructs with which the philosopher describes and discriminates (Proposition 10).

As we saw. Sartre rejects thing-like 'neutral givens', both 'in the psyche' and 'in the world'. Phenomenologists assume no static 'objects' or variables of observation 'out there' or representations of them 'in here'. Hence observation is no mere gaping at [5] already given entities, variables or perceptions which we need only 'associate' (Proposition 9).

Activity-in-context replaces the view of given entities. Why not, then, fashion philosophic terms that are not static representations?

We must not call Sartre's type of term 'ambiguous' if we are [Page 153] bothered only because it isn't a representational type of term. For this nonrepresentational character of human activity Sartre employs paired terms, designed for use in a movement from each to the other. Moving between these dialectically paired terms often has a paradoxical sound. Linguistic analysts should not miss their kinship to this mode of philosophizing because of a dislike for such terms. After all, linguistic analysts have not yet fashioned terms for this nonrepresentational aspect so central also to their method. Linguistic analysts bring home this same aspect, but they do it with examples—and these are also intentionally paradoxical, to point up the fact that activity in situations is not reducible to units, pictures, objects of knowledge, representations.

Both methods reject reasoning from constructs alone, and wish to evaluate constructs only in terms of what is done with them, how they are used, what they directly discriminate. It follows that we should not turn away from the type of constructs used before seeing what their use discriminates.

Sartre's dialectical terms do not make an abstract dialectical scheme. Traditional dialectitians object to Sartre's 'dialectic' because it remains on the 'first level' on which it starts. Sartre refuses to 'raise' his contradictory terms to a 'higher' (more abstract) synthesis. For Sartre, when Hegel's "being" and "nothingness" turn into each other and are then absorbed in a higher synthesis, they become "mere concepts" (BN, p. 16)

A pair of Sartre's terms are never absorbed. In moving from one to the other and back we discriminate an aspect of concrete living and moving, that cannot be represented by single static representations.

Such a pair of terms gets at something Sartre finds in a great many aspects of human living: "scissiparity" (a word reminiscent of scissors, borrowed from biology where it is used for amoeboid splitting of one organism into many). As Sartre uses it, 'scissiparity' names something he sees over and over again: some facet of human activity which is really one, and yet appears to have two poles. The traditional misrepresentation of the activity represents the two poles as though they were entities or things, and the activity is thereby misconstrued or lost. Thus consciousness (an activity) always seemingly involves "something present to [Page 154] something," and there you have the temptation to describe it all in terms of thing-like contents in a mirror-like consciousness.

For example: "reflected-reflecting." When we reflect on our own consciousness, to be conscious "of" always involves also our ongoing process or activity. We must not take one term as just itself ("it isn't identical with itself," Sartre will say). Nor can we simply put both terms together, as two representations tied together. That gives only the poles, not the movement. The wish to have the totality represented must fail. The movement, the activity, the process, is being discriminated.

Thus by the sole fact that my belief is apprehended as belief, it is no longer only belief: that is, it is already no longer belief, it is troubled belief. Thus the . . . judgment 'belief is consciousness (of) belief' can under no circumstances be taken as a statement of identity (BN, p.75).

The phrase "no longer" illustrates the movement to which Sartre points with the pair "reflected-reflecting." These terms make no sense considered as separable somethings (as each "identical to itself"):

On no account can we say that consciousness is consciousness or that belief is belief. Each of these terms refers to the other and passes into the other, and yet each term is different from the other. We have seen that neither belief nor pleasure nor joy can exist before being conscious . . . (BN, p.75).

Sartre offers many other such pairs and with them he can characterize many detailed aspects of living, loving, smoking, doing, making, having, and so on. (Like linguistic analysis phenomenology seems unphilosophical to some because of its many new discriminations of life detail.)

The pairs of terms avoid the pitfall of a representational analysis. But are not these pairs themselves representations? No, they are not. Are they the proscribed "laws of consciousness"? No, they are not. Applying the method to our own use of it, we have to accept that we use not what the terms picture but our own on-[Page 155]going activity. This activity is not being represented by the terms. They 'point to' this activity (much as examples in Linguistic Analysis do). Thus these philosophic terms are not 'the' structure of consciousness, but 'pointers' which newly discriminate concretely had aspects of what we do, have, notice. We ('implicitly') have and know what the terms get at, point to and explicate. (Otherwise they would be again just another set of constructs, "laws" or "representations" of consciousness.) (Proposition 8).

Our own concretely ongoing activity 'sustains' concepts or laws or representations or structures or terms. It is always more basic than they and not made up out of them.

. . . consciousness is not produced as a particular instance of an abstract possibility but . . . it creates and supports its essence—that is, the synthetic order of its possibilities (BN, p. lv).

These 'possibilities' of action constitute the 'self'. We seem to have a 'self', 'interrogate ourselves', 'refer to' ourselves inwardly, and 'aim' questions at ourselves. But Sartre denies that 'contents', 'entities' or 'meanings' characterize consciousness.

. . . some wrongly hold . . . the 'I' . . . to be the inhabitant of consciousness.
. . . through hypostasizing the being of the . . . reflected-on . . . these writers fix and destroy the movement of reflection upon the self . . . We on the contrary, have shown that the self on principle cannot inhabit consciousness."
It is an absent-present (note how this pair of terms operates) . . . the existence of reference . . . is clearly marked . . . (Consciousness refers) down there, beyond its grasp, in the far reaches of its possibilities (BN, p. 103).

We refer not to something actually there but to something "absent," the "far reaches of its possibilities."

But this possible . . . is not present as an object . . . for in that case it would be reflected-on (BN, p. 104).

Most often we talk of the 'self' as something unquestionably 'present' inside us—which misses the peculiar way in which we must interrogate, seemingly to find out what we are, feel, etc.

The Freudian 'ego' and 'id' come directly from this interrogating and digging. Again the poles of the activity are made into [Page 156] entities: the ego; the id. Sartre, phenomenologically, rejects both the constructs of a present and of an absent entity. It is the 'absent-present', not as a paradoxical sticking together of both constructs, but as a delineation of what we do: we seem to 'refer' to inward contents but—we find: for example, I am thirsty. What do I 'refer to'? My 'thirst' as a content, like 'pleasure', an object 'of' which I am conscious? Rather, it is the 'possible' of drinking:

Actions in situations are 'implicit' in the way consciousness "refers" to itself-qua "absent":

But this possible . . . is non-thetically (not like an object is given) an absent-present . . . The satisfied thirst which haunts my actual thirst (it "haunts": it is not baldly here nor not here, it is my desire or possibility) is consciousness of itself-drinking-from-a-glass and a non-positional consciousness.

Rather than 'a thirst' as a 'content', when we 'refer' to and 'interrogate' our 'absent-present' self, we find the possibility of drinking from a glass—and this possibility of action in the situation is not given 'in there' as an object of reflection. Rather, there is this 'itself-drinking-from-a-glass' and a 'nonpositional consciousness' (the last phrase reminds us that it is concrete activity. If it were not, we would again interpret activity as representations, this time as representations of possible actions).

Actions in the situation (our possibilities) are 'implicit' in this activity of 'referring' to ourselves as 'down there' in our feelings. Indeed, every explication of feelings always reveals not entities like anger, fear, thirst, pleasure, etc., but anger at this and this situation, because it forces us to do so and so, and give up such and such, or fight with so and so. These are what we might do in the situation (Proposition 7).

How is it that we 'refer' to what we bodily feel and thereby 'make to be' such possibilities of situational action? The body is the "condition of possibility" (BN, p. 338). We are bodily in situations. The ongoing activity of consciousness is this bodily being in the world. This bodily feel Sartre calls "nausea" but he indicates that it needn't have just that specific quality. It is your ever present live sentient feel. He also calls it "coenesthetic affectivity." It is the body, again not as entities but as activity in context. That is what we 'interrogate' when we 'aim' questions 'at' [Page 157] our self 'down there'; the 'feeling' which isn't a content but the 'possibilities' of action:

Coenesthetic affectivity . . . provides the implicit matter of all the phenomena of the psyche . . . it is this which we aim at . . . and form into images . . . in order to aim at absent feelings and make them present . . . (BN, p. 338).

And now, the problem of criteria: Granted that we are looking 'directly', without the 'blinkers' of words or constructs, and granting that a good 'description' leads us to notice newly discriminate aspects, nevertheless: do our descriptions not import a variety of assumption systems each with different consequences?

Consider that Merleau-Ponty—with similarly phenomenological intentions—argues strongly for a very different type of term and different assertions.

We have to grant that while philosophic statements 'lift out' what is 'implicit', 'already there', 'given' or 'noticeable' directly, it is not there in the structured units and patterns descriptions impose (Proposition 6).

These philosophers emphasize that 'phenomena' or activities ('consciousness', 'perception', 'being in the world') are not constituted of given units, things, representations, meanings, etc. Thus it is not the question whether Sartre's or Merleau-Ponty's descriptive terms represent the really 'given' units and structures. Both emphasize the 'pre-reflective' and 'prethematic' character of living activity (Proposition 5).

It follows that we must look more closely at how phenomenological descriptions are 'based on' our direct looking and living. Obviously, 'based on' here does not mark a correspondence of structure such that the formulation's structure corresponds to the structure of the given (although, why oh why is just that so often claimed?)

When a phenomenological philosopher offers his descriptions as 'ontology' he claims only that we already know 'pre-ontologically' everything he has to say. We know it in living, but without the explicit structure which he 'lifts out' with his description (Proposition 4).

But, if descriptions are not congruent with the pattern already there (because what is there isn't given so patterned), then are [Page 158] they arbitrarily imposed? Are phenomenological philosophies mere speculative assumption systems no different from other such systems? Is the claim to phenomenological grounding merely an unfounded claim to special privilege for one's assumptions?

Just as linguistic analysts feel you are missing the point when you question the constructs and assumptions in their "descriptions" (of circumstances), so phenomenologists feel that the point is being missed when the variety of construct systems is questioned. The point is that as a result of a description (no matter how wild its terms) you may directly notice something you had not previously noticed. Do not evaluate the description as a set of constructs, but as a set of pointers. All constructs are evaluated by what they discriminate directly (Proposition 3).

But suppose we do not miss the point. Suppose we call a description 'good' whenever it 'lifts out' something for us. Even then, we must ask: can two different descriptions lift out the 'same' aspect for our notice? If we say yes, we assume that the given activities come already cut up (such that there is a 'same' aspect waiting there apart from the differing descriptions). If we say no, we admit that the supposed directly discriminated aspects are really a function of our descriptive 'blinkers'.

We must view phenomenological explicating also as an activity. Then we will not assume that it is determined by some system of entities or constructs that lies beneath phenomena. Hence no 'same' aspects can be waiting for us there (Proposition 2).

But our activity (as much during explicating as any other time) doesn't only 'sustain' representations (it isn't reducible to representations). It also 'surpasses' them . . . . it reorganizes, reinterprets, creates new alternatives, new possibilities. For example, Sartre does not believe in an abstract 'freedom' such that I might be able to leap out of the situation I am in. (If I am a cafe waiter, this role doesn't define me. It is a representation which I sustain, but I cannot suddenly be a diplomat instead.) Activity surpasses representations in all sorts of ways, but not in just any old way: always in regard to the situation I am in: the 'facticity' of the situation (Proposition 1).

The pair of terms: 'facticity-surpassing' is another Sartrian pair. A situation's factual constraints cannot be described apart from my activity and possibilities. Its factual constraints are created [Page 159] by, posed for, and in terms of, my possible activities in the situation. Similarly, my activities always create new possibilities (and thereby aspects of the situation which it couldn't have been said to have, before and apart from me). Similarly, the world isn't given in just such and such a structure so that we might read it off. Rather, our activity creates, sustains and surpasses the patterns we explicate.

But it is clear, therefore, that we lack any way to examine the various explications, how they newly 'make be', 'surpass', 'split up', and describe, and their various unexamined assumption systems and consequences.

We cannot leave this problem in this shape: all claims of a phenomenologist's base for description is in question. We must ask: what then is the way in which explication is 'based' the implicit pre-structured?

III

The 'implicit' factor is so central in these methods, that we have already had to say a good deal about it, and about 'explication'. The propositions I offered constitute a treatment of explication, provided we now continue.

(a) We said that a 'good' explication statement (we were asking for criteria for such statements) leads us to notice directly some aspect we had not previously noticed. That is actually a striking criterion (of a peculiar sort, to be sure) which sets successful 'explication statements' apart from the many statements one can always formulate which do not succeed in bringing something new to direct notice.

(b) Furthermore, we don't always call a statement an 'explication' when it leads us to a newly noticed aspect. One can always devise very many statements which (by objective criteria) state the many facts that were there and had not been noticed. One does not call these 'explications'. It is an 'explication' only if we are 'sure' that the new aspect 'was implicit' before (i.e., we now insist that it was 'known' to us, or 'there' for us). This retrospective assertion of the newly noticed aspect is also something only few special statements bring about.

The new aspect (once noticed) must have this special relation-[Page 160]ship to what we do remember noticing (or feeling, or knowing) so that we now 'insist' it 'was implicit' in what we knew, felt, or noticed, and not just another fact in the situation.

(c) Once an aspect is discriminated and newly noticed, it cannot be made unnoticeable again. Of course, one need not pay attention to it, one can forget it, consider something else, etc. But once discriminated, the aspect cannot directly be made to merge away again.

(d) These three powers of a newly discriminated aspect are not totally dependent on the explication statement. The statement leads us to the newly discriminated aspect, but once it has done so, the aspect is noticed 'directly'. Also its quality of 'having been implicit' in what we knew or did before doesn't come from the statement (usually cannot be reduced to formal relationships between statements of what we knew before and this statement). We may throw the statement out and still the newly discriminated aspect cannot thereby be made to disappear again.

Because of this partial 'independence' of the discriminated aspect we can say that there is 'direct' noticing and not only various statements. If what is directly noticed depended entirely on the statement (not only for its being noticed, but also for its quality and for its remaining noticeable) then all discussions would again become a matter of various formulations related to, or clashing with, each other.

(e) But, neither can we say that the newly discriminated aspect is fully independent of the variety of statements once we notice it. It would be convenient (though, in the long run it wouldn't be at all desirable) if we could say flatly that once we notice the aspect we can use various statements of it equivalently for this 'same' aspect. At a particular juncture of some discussion or tasks two very different statements might serve to discriminate the 'same' aspect, but the very next step of the discussion, or the next difficulty in the task, might require that we discriminate further aspects, and we may then note that the two statements are no longer equivalent (or, perhaps they still are—but we must look).

(f) Because the newly discriminated aspect is the function and purpose of the statement, we need not argue from the statement as from a model: we can but we need not be bound by the logical necessity with which all kinds of logical implications follow from the statement.

[Page 161]

Yet, if we were to deny the statement all logical character it would cease to have any discriminating power. How do we decide what logical implications we use, as against those we ignore? We decide by noticing what the logical implication may further discriminate in the directly noticeable aspect.

This means that logically different explication statements with different models can always discriminate different further aspects of anything we directly notice. It is a decision, to consider these differing further aspects unimportant (for the present discussion, task, etc.)—a decision which has to be made at every step, since at each step these differing aspects may become important.

Thus the logical characters of different explications can neither be always accepted nor always ignored, but must be pursued to notice what is differently discriminated by them. Only so can we have explication statements that are not again mere formulative assumptions.

(g) I cannot go into it very far here, but because of the way one can neither drop, nor uncritically accept the logical structural aspects of explication statements, there is a whole field of such procedural choices we use and must examine. [6]

Either linguistic analysis and phenomenology are to be considered merely arbitrary play with arbitrarily chosen assumption systems, arbitrary ways of selecting and defining new distinctions, arbitrary imposition of formulative patterns, that is to say, either we consider these modes of philosophy to be no different (though less self-critical) than traditional philosophy, or we will have to grant the central role of directly had, not yet formulated experiencing.

Only because these methods involve the use, during philosophizing, of directly had, not yet formulated experiencing, is the appeal to phenomena or to 'direct looking' more than a circular and invalid claim to special privilege.

But this raises, as a central problem, the question of how formulations may be related to directly had, not yet formulated experiencing.

Sometimes this directly had, not yet formulated experiencing [Page 162] is called 'feeling'; (for example, in this common conversation: Question: "On what do you base your assertion that this rule explicates the use of this word?" Answer: "The use it states feels right to me. Doesn't it feel right to you also?"). Either we take such assertions of 'feeling' to be merely self-righteous claims, or we must really examine what sort of relationships a formulation may have to 'feeling'. (You might call it 'sensing': "Ryle sensed trouble where trouble was" says Cavell [7]).

Notice also, that such 'feeling', 'sensing', or not yet formulated but directly had experiencing is used as we formulate. Similarly, the few criteria and procedural distinctions given above concern the use of direct feeling or sensing (We 'insist' . . . we are 'sure' we have a direct sense that the newly discriminated aspect was involved in what we did notice earlier . . .). We can further explicate and give a logical rendition of any such instance, but this still involves the direct use of the experiential discrimination so that we may explicate it and thereby set up logical explicit accounts of how we have already proceeded.

Philosophy is not the only discipline in which this relationship between the directly felt experiencing and conceptualization arises. Another such discipline is psychotherapy. [8] There, too, an individual freshly discriminates and conceptualizes directly felt, not yet formulated experiencing. There, too, the individual backs his formulations up with insistence that it 'feels right', and unless this be fatuous, much more is involved in this peculiar relationship between direct feeling and concepts.

And, does it not seem now that these philosophic approaches have after all fallen back on a 'private datum' or 'content', much as they eschew such a view? Again, this will be the case only if we leave the relation between feeling and concepts unexamined. If we examine it, we notice that we deny the assumption that the whole sequence of explication (or behavior) is somehow folded into the feeling, like an accordion. We deny that feeling, as such, needs to be 'checked' against, (as in the spurious problem of the 'private' supposed basis for first person statements). We have [Page 163] already seen that feelings or 'implicit' knowledge is not such as to permit checking its correspondence with what is said. Even if the feeling were not at all 'private', even if it were an object on the table, it would not contain in itself explicitly the whole sequence of explications or behaviors. Explication and behavior occur in the world. Feeling ('implicit' experiencing) is not fully behaved, not fully explicated. As not fully behaved it does not yet contain the objective sequence. But it functions centrally (we must see how) in behaving or explicating, and when it does, there is no problem of observing its bearing in the world. For example, my six year old daughter, just before getting the mumps, had a pain and told me about it in an explicated and behaved way which needed no 'checking': "Daddy, right here under my ear (pointing) it feels like a black-and-blue mark, but I looked in the mirror and there isn't any black-and-blue mark there!"

When not-yet-formulated experiencing plays its role in formulating and behaving, the way it bears on aspects of the world becomes quite clear and observable. Before it plays such a role, it is not yet all that. The activity of explicating or behaving doesn't just play out steps fully contained in pre-formulated experiencing. Rather, it carries experiencing forward, formulates and creatively shapes it further; and yet a given feeling won't function to support just any and all formulative attempts (as our 'criteria' above show). Only some very fortunate formulations (often we fail to find any) obtain these characteristic responses from the directly felt.

Thus, on the one hand, the directly experienced 'implicit' knowledge we feel doesn't fully determine the formulations and hence leaves us open to assumptions and contradictory possibilities. On the other hand, it doesn't permit any and all arbitrary formulations. It is therefore essential that we specify the kinds of support which the directly used 'implicit' gives to formulations.

Such an examination must occur in terms of procedural choices which can be specified. The role of feeling in explicating can't be examined except in explicating. But, there it must be examined, otherwise very different procedures in regard to formulational models will be mixed together, our use of them will be arbitrary, our accepting some logical implications of some models will be arbitrary and our rejection of other logical implications of the same models will also be arbitrary, we will have no way of dealing [Page 164] with contradictions between seemingly applicable 'rules' or 'explications', and no systematic way of appealing beyond mere theoretical constructs to direct experience.

Elsewhere [9], I have attempted this examination. Here I have been able to give only a few instances.

My main purpose in this paper has been to show that both phenomenology and linguistic analysis employ net yet formulated, directly had experiencing in their methods. I have tried to show that this is no abstract or ephemeral thing, but something the individual must directly have and use, otherwise the rules and explications he formulates are merely arbitrary.

I have tried to show that this type of philosophy indeed breaks out beyond mere theoretical assumptions and constructs, that it breaks out toward directly had and used 'implicit' experience, not yet formulated or cut up into neatly packaged traits, variables, denotations, etc.

The vicious circle I pointed out was not intended to lead us back to the imposition (the reading in) of theoretical constructs as basic immutable assumptions. Instead, the problem leads forward to an examination of the relationships between the directly sensed 'implicit' experiencing we use, and the variety of formulations and procedural choices we make.

Eugene T. Gendlin
University of Chicago