Please come back when you're done mixing up categories. — StreetlightX

If you don't want to discuss the text with me, that's fine. But to make such unsupported accusations is not. Clearly, as per the quotes provided by me above, what is being described in this chapter of the book is "absolute ideality". If there is any "mixing up categories" going on here, it is being done by the one (Derrida) who claims that the possibility of repetition is a necessary condition of absolute ideality.

If you would consider the first page of the chapter, p41, it is described how, for Husserl, (2) internal discourse is not a case of communicating to myself, "...the existence of psychical acts is immediately present to the subject in the present instant."

Therefore it is impossible that this internal discourse is understood as a "representation". "Immediately present" denies the possibility of representation, which indicates that the object being considered is not immediately present.

Further, it is explained in the quote from Husserl, "that one merely represents oneself as speaking and communicating". So according to Husserl, we must distinguish between what is actually occurring in the case of internal discourse, from the representation of this, which is a speaking and communicating. Then Derrida explains that to avoid this distinction between what is really happening (reality), and the representation of it, we would have to follow Husserl into the category of "fiction" which Husserl defines as a "neutralizing representation".

So instead of going into this concept of "neutralizing representation", Derrida claims that it is impossible, in practise, to make such a distinction, between reality and representation, and that this impossibility is not produced in language, language is that impossibility. This is where Derrida makes the mistake which you accuse me of, mixing up categories. What is being discussed is a separation between reality and representation within ones own mind, and this is necessarily a theoretical separation, an ideal. By asserting the impossibility of this theoretical division, in practise, Derrida finds reason to move from the category of the ideal, to the category of practise, which is other than ideal, and proceeds to discuss the properties of language, as they occur in the practise of communication. But this claim of "impossible" is unsupported

The first mistake which manifests at the bottom of p42, is the claim that when I make use of words, I must do so in "a structure of repetition whose element can only be representation". This is clearly false. It is only by limiting the existence of the word, to being a property of communicative language, that such a conclusion follows. As I indicated earlier, the example of music gives us repetitive sounds, words, syllables and tones, which are often not meant to represent. We use them to entertain, stimulate us, bringing us passion and spirit, rather than representation.

The possibility of such reality, the reality of music and other art forms, which is not representative at all, indicates that Derrida's claim that it is impossible to distinguish between reality and representation, and that the use of the imagination is necessarily representative, is not accurate. Perhaps even some metaphor may be free from representation. Clearly we can imagine a reality which excludes representation. But Derrida appears to proceed from this false premise of "impossibility", that there cannot be expression which is not in some way representative.

And it's this strange, confused knot in Husserl's text that Derrida uses as a springboard to jump into *all* the contradictions and hidden assumptions of phenomenology. — csalisbury

Before we jump on that springboard we should consider it for soundness, and examine it for weakness.

I may just be more sympathetic to Husserl, but I don't find expression more mysterious than indication. It's linguistic meaning capable of taking part in logical relations, as intended by some agent. The soliloquy example itself also doesn't seem strange to me, although it's a separate question how strange it is after Derrida's criticism, viz. that imaginary and actual uses of language are both equally indicative.

MU, imagination is representative not in the sense that it has to depict some real thing, but in the sense that for Husserl it's derivative of perception, which is presentation proper. Imagination and fiction aren't just ideal: when fantasizing, there's a concrete act going on as well, and a particular fantasied object. — The Great Whatever

Yes, sure there is a concrete act occurring as the act of fantasizing, but is it not the fantasized object which is immediately present to the mind of the fantasizer rather than the act of fantasizing? If the object is immediately present, then how is this a representation? It must be a presentation of the object rather than a representation.

Regardless of whether expression is mysterious, though, Husserl treats it mysteriously. My impression of the structure of 'essential distinctions' in LI is that it runs something like this: Husserl distinguishes between indication and expression, then slowly, step by step, excises indicative elements until only expression is left - the end-product is the chapter on soliloquy. It's very much like 'reduction' in the sense that street mentioned earlier in this thread, not an 'epoche,' but a reduction in the sense that one 'reduces' a sauce. Only after this reduction, does he finally move on to a broader discussion of sense and reference. But it seems strange that the one situation Husserl identifies in which expression is fully, exclusively, expression is one in which expression is entirely superfluous. If we take the idea of expression in Ideas, where an inner sense is 'exteriorized' in a sign, then we have a situation in which someone is externalizing and instantly reabsorbing something he already has in full.

It's not entirely clear. Husserl thinks that there's a sense in which perception is primary presentation, and imagination, memory, etc. are secondary, hence re-presentation. So it's reproductive without actually having to reproduce some really perceived thing – every fantasy is (merely) a fantasy of some conceivable perception. And of course there's the decreased vivacity, the need for the imaginer to actively keep the fantasy in existence.

Maybe the reason I'm not that surprised by this is that I think I can and have done it? Whether or not this ability is independently mysterious, it seems to me I can in soliloquy describe what I see with a self-present intention in such a way that I'm not really 'telling myself' anything at all in the sense of giving myself new information I didn't know or intimating my own thoughts to myself, but just expressing what something is in silent words appropriate to a perception. And I think in doing so, yeah, I can 'see' what the words mean, which perceptions they'd be appropriate to and which not, and so on. It seems commonplace to me.

yeah, i guess I wasn't looking at as something that simple or commonplace, but that makes sense - and you're right, it's eminently doable. But now I feel adrift again in terms of understanding V&P. Like the actual/imagined distinction w/r/t words would be simply an individual utterance vs the word itself considered ideally. Which is totally valid, even if iterability is baked into the sign.

I see the pull of the idea that a sign's function is carried out in full whether any actual tokening of the sign exists, and that even an imaginary sign defers to a symbolic system in the same way it does in actual communication. I'm still a little shaky on what this means or how it corrupts all expression with indication, unless the point is something like 'a sign always indicates the symbolic system it belongs to and all of its other possible uses' which doesn't seem right.

I've always liked Merleau-Ponty's take on this, who provides an account of the transcendental illusion (or what he calls the 'trick') involved in thinking that thought is 'inner': "Thought is nothing “inner,” nor does it exist outside the world and outside of words. What tricks us here, what makes us believe in a thought that could exist for itself prior to expression, are the already constituted and already expressed thoughts that we can silently recall to ourselves and by which we give ourselves the illusion of an inner life. But in fact, this supposed silence is buzzing with words – this inner life is an inner language. “Pure” thought is reduced to a certain emptiness of consciousness and to an instantaneous desire. The new meaningful intention only knows itself by donning already available significations, which are the results of previous acts of expression" (PoP, p. 188-189, Donald Landes trans.)

*M-P uses the term 'expression' here in a different way than the technical sense we've been discussing here; he means it in the more colloquial sense of 'to express an inner thought' - although this is just what he will challenge.

There are some incredible resonances between Merleau-Ponty's and Derrida's work on language, although in truth, I much prefer reading M-P than Derrida. There's just something... elating about M-P's use of language, where Derrida comes off as just trying to be a bit too clever alot of the time.

What tricks us here, what makes us believe in a thought that could exist for itself prior to expression, are the already constituted and already expressed thoughts that we can silently recall to ourselves and by which we give ourselves the illusion of an inner life. — StreetlightX

Out of curiosity, do you believe this? It seems like it can give you a sort of theoretical elegance, especially if, like MP, and Derrida, you want thought and experience to have a structuralist flavor (and MP's increasing dependence on structuralism is a little questionable). But it seems to be straightforwardly wrong phenomenologically, and no insisting on the contrary is really going to help.

Why is it considered such a philosophical virtue to everywhere deflate or otherwise 'expose' the notion of something being inner? Is it because philosophers only know how to deal with the external?

It's not entirely clear. — The Great Whatever

Well, it seems to be a pivotal point. If there is ambiguity on Husserl's part, then although StreetlightX has accused me of category error, and I have counter-accused Derrida of the same mistake, it could be that we each have equally valid interpretations. But these interpretations appear to be vastly different.

Let me see if I can sort this out. In chapter three we had the distinction between the imagined word, and the act of imagining. Now at the beginning of chapter four we have a distinction between the act of imaging words, and the representation of this, "speaking to oneself". We can take the act of imagining words as what is real, and the "speaking to oneself" as a representation of this.

Now, at p42 Derrida makes the claim that in language, we cannot distinguish "rigorously" the difference between what is real, and what is representative, so this is referred to as an "impossibility". What I believe is relevant at this point is the distinction between the real act of imagining words, and the representation of this, "speaking to oneself". Surely we can distinguish between the act of imagining, and the representation of this, the description, "speaking to oneself".

I've tried and I've tried, but I cannot understand the reasoning for Derrida's claim of this impossibility. Perhaps StreetlightX can explain why Derrida is insistent on this claim. First it is claimed, "Between actual communication (indication) and 'represented' communication, there would be an essential difference, a simple exteriority." But this is clearly a misrepresentation of Husserl's stated position, that "speaking to oneself" is the representation. The imagining words is not a communication at all, it is only represented as communication, "speaking to oneself". So this is not a case of actual communication and represented communication as Derrida presents it, it is a case of imagining words, which is not supposed to be a form of communication at all, being represented as a form of communication, speaking to oneself.

Derrida then proceeds to talk about the "actual" practise of language, but as far as I can tell, this is a reversal of Husserl's position. Husserl has exposed something, imagining words, which is actually not a practise of language, it is only represented as a practise of language, "speaking to oneself" and Derrida now treats this as if it is an actual practise of using language, and proceeds with his argument. Now I really do not see how it is claimed that we cannot distinguish between this thing, imagining words, and the representation of it "speaking to oneself", such that we would believe that the thing represented, imagining words, is an actual practise of language.

I'm still a little shaky on what this means or how it corrupts all expression with indication, unless the point is something like 'a sign always indicates the symbolic system it belongs to and all of its other possible uses' which doesn't seem right.

I don't know if that is the point, but it does seem plausible me that a sign always indicates --maybe not the entirety of the symbolic system to which it belongs, or all of its other possible uses - but that it indicates at least the linguistic neighborhood of which it is part. I suppose you could say the sign always indicates a larger language game (without necessary saying it indicates the totality of the system of signification)

To add to that, you'd mentioned, much earlier in this thread, sentences used to teach another language. Barthes has an interesting passage on this in his Mythologies

"I shall borrow [an example] from an observation by Valery. I am a pupil in the second form in a French lycee. I open my Latin grammar, and I read a sentence, borrowed from Aesop or Phaedrus: quia ego nominor leo. I stop and think. There is something ambiguous about this statement: on the one hand, the words in it do have a simple meaning: because my name is lion. And on the other hand, the sentence is evidently there in order to signify something else to me. Inasmuch as it is addressed to me, a pupil in the second form, it tells me clearly: I am a grammatical example meant to illustrate the rule about the agreement of the predicate. I am even forced to realize that the sentence in no way signifies its meaning to me, that it tries very little to tell me something about the lion and what sort of name he has; its true and fundamental signification is to impose itself on me as the presence of a certain agreement of the predicate. I conclude that I am faced with a particular, greater, semiological system, since it is co-extensive with the language: there is, indeed, a signifier, but this signifier is itself formed by a sum of signs, it is in itself a first semiological system (my name is lion). Thereafter, the formal pattern is correctly unfolded: there is a signified (I am a grammatical example and there is a global signification, which is none other than the correlation of the signifier and the signified; for neither the naming of the lion nor the grammatical example are given separately."

I reread Chapter 4 this afternoon and don't really understand any better. The point about language use in imagination and actuality collapsing because it is representative equally in each case, and that the distinction between reality and representation breaks down there, is on the cusp of making some kind of sense, but I still just can't see why language is supposed to be shot through with fiction, or however you want to put it.

The tensions between being as ideality (infinite repeatability) and presence (full presence with no need for repetition) makes a little more sense, though I now agree with you that the reference to mortality could only be plausible for the end of the form of the present in general, not for me personally. Though maybe since the form of the present is linked to the transcendental ego, Derrida is hinting at the deeper foundation of the form of 'now' within myself.

but I still just can't see why language is supposed to be shot through with fiction, or however you want to put it.

Just spitballing here, but maybe the idea is that, since the word is ideal in both actual and fictional discourse, language functions the same way whether used nonfictionally or fictionally. Like, that we are able to write fictions at all is because language operates in this ideal, iterable way. Words don't simply correspond immediately with their referents but operate according to this ideal iterability.

Does that make sense ?

I agree with that as far as writing fiction as a genre. I thought the focus was on fiction in the sense of imagining speech (whether the imaginary speech is intended to be fictional or non-fictional in genre).

I feel like I almost get it, but it's just not clicking for me. Isn't Husserl going to agree the ideal function is retained in imagination? Isn't that the point of doing expressive exercises in soliloquy? What would be a problem is if language's indicative functions follow Husserl into soliloquy. So Derrida's move requires something more radical – linguistic signs always have the same functions, imagined or not: actual linguistic use is 'representative' as much as supposedly 'representational' (imagined) use is actual. You can't 'imagine' a discourse because to imagine it is just to have it.

What I believe is relevant at this point is the distinction between the real act of imagining words, and the representation of this, "speaking to oneself". — Metaphysician Undercover

I think it is the difference between a real communicative linguistic act and an imagination of this (which is the representation). Not a a representation of the imagination, which goes one level too far.

Derrida then proceeds to talk about the "actual" practise of language, but as far as I can tell, this is a reversal of Husserl's position. Husserl has exposed something, imagining words, which is actually not a practise of language, it is only represented as a practise of language, "speaking to oneself" and Derrida now treats this as if it is an actual practise of using language, and proceeds with his argument. Now I really do not see how it is claimed that we cannot distinguish between this thing, imagining words, and the representation of it "speaking to oneself", such that we would believe that the thing represented, imagining words, is an actual practise of language. — Metaphysician Undercover

But yeah, this sounds right. We need to understand what it is about linguistic signs that makes Derrida think their being used, and imagination of their usage, collapses.

You can't 'imagine' a discourse because to imagine it is just to have it.

Yeah, and that actually makes sense to me intuitively. I suppose the next step would be that if imagined and actual discourse are the same thing, by imagining discourse one introduces the same play of absence/presence reality/representation you find in communication into the very self-presence of the imaginer. The problem is that discourse is always working no matter where and how it appears, so you can't do the thing of calling it an 'irreal' product of a noetic act in order to fix it in place and observe it. The category of 'existence' never applied to it in the first place, and, as discourse, it continually produces itself. It kind of fuck ups the unilateral constitutor-constituted thing. I guess, that is, you can't observe discourse without actually being drawn into it and, in a way, doing it.

I'm not sure how much sense that makes. I feel similar to you, it's all kind of there in a tip-of-the-tonguey way, without quite clicking - It also gives me that slightly nauseatingly recursive feeling I get when trying to do something like e.g. determine if the word 'autological' is itself autological.

I think it is the difference between a real communicative linguistic act and an imagination of this (which is the representation). Not a a representation of the imagination, which goes one level too far. — The Great Whatever

The problem I see with this is that in the quote from Husserl, on p41, it is clearly stated: "...there is no speech in such cases, nor does one tell oneself anything; one merely represents oneself (man stellt sich vor) as speaking and communicating."

So there is an activity, which has been referred to as imagining words, and this is represented as "speaking to oneself". The act of imagining words is that act in which there is no actual speech, nor does one really tell oneself anything, but this is represented as a speaking or communicating with oneself.

Out of curiosity, do you believe this? — The Great Whatever

Yes, with qualification. At this point in his writing, thought remains a matter of 'words' for M-P (i.e. the first line you didn't quote: "Thought is nothing “inner,” nor does it exist outside the world and outside of words). But as M-P would realize (and this is where Derrida more or less 'begins'), 'words' are a case of a more general phenomenon of what he calls 'expression' (employed in a very different way than in Husserl). Expression is perhaps the the most interesting concept in all of MP, and it designates a paradox in which what is expressed does not pre-exist it's expression, but is engendered along with it. For MP, expression will become the manner in which pretty much every phenomenon is subject to, and comes about from (from perception to language, art and ontology).

So I would shift the emphasis on 'language' and 'words' in the passage, to expression tout court. Although already couched in the language of expression, at this point, language remains the model of expression in MP. But if - in a manner analogous to Derrida - expression becomes a model unto itself as it were, one that applies to all phenomena indiscriminately, then thought is just as much a matter of say, gesture, as it is words (gesture being another theme that is incredibly important in MP). One can literally think in gestures, in movement: thought as affect. So there's something more to 'inner thought' than the mere recollection of expressed language or expressed speech. Insofar as language itself is - as far as I'm concerned - a species of affect, what is 'silently recalled' is more or less a complex of affects, sometimes couched in the form of language, sometimes not.

To try and bring this back a little to Derrida, the point of convergence is that just as M-P generalizes 'expression' (initially a feature of language, now a feature of things more generally), so too does Derrida generalize the 'structure of repetition' as no longer belonging solely to the sign, but as 'contaminating' any and all 'presence'. In both cases, what is attacked is any kind of 'simple' from which everything else flows: inner speech, presence, etc. In both cases it's always a matter of attending to the reality of expression itself, of bestowing existential weight, as it were, the process of unfolding, rather than treating expression (in MP's sense) as immaterial scaffolding. They are both 'anti-reductionist' imperatives, where things can't be 'reduced' to other things without remainder. So it's not a matter of siding with an 'outside' against an 'inside', so much as refusing any attempt to isolate any 'region' whatsoever as being somehow primary, with everything else derivitive.

This is why Derrida will end up talking about 'originitive supplements' at the end of VP, where supplementarity (derivation) is mapped right into the 'original' itself. But we'll get there soon enough.

I think I agree, all I am saying is that sometimes the way you are wording it makes it seem as if there's first an imagination of a word, and then a representation of that imagination. What I'm saying is that's one level removed, and the text treats the imagination as the representation to begin with (of the actual use of the speech in communication).

Okay. All I am saying is that I don't see anything of thought as I actually live through it in these accounts. It sounds like an alien who lives a very different life from me describing the way they think, because I don't think that way. The folk description of thought and inner life is much more accurate.

I don't think my inner life is primary, merely that it is primary for me. And that it is distinct from my outer life and not reducible to it, or to gestures and language, and the majority of it is incommunicable.

I definitely don't have a voice that exhaustively narrates everything I experience, but I do find, if I pay close attention, that there's usually a sort of murmur. I'd have trouble agreeing that my inner life isn't inner, but it's definitely laced through with (and built deeply upon) layer upon layer of inherited forms.

I agree, but I don't see any reason to believe that those inherited forms privilege language or gesture in any interesting way. Even trying to conceptualize what that would be like is difficult. I imagine myself playing a text adventure game or Dance Dance Revolution; a caricature of living.

My inherited forms seem to me something more like a soul or personality, a memory in every novelty, which is refracted through a prism that is 'me.' In other words, I just can't agree with Hume: when I look into myself I do find myself, and his description seems to be one of someone who is very, very high looking through a kaleidoscope. Furthermore, my skepticism toward these positions increases insofar as I am attending to how I actually live, and decreases insofar as I am attendant to resolving aporia in the philosophy literature.

Not to get too deep into it, and get too off topic, but when I look into myself, I find...well yes a soul, or personality, but it's something like - if I focus deeply, or if I'm high - it's like a collection of many 'choices' (though the term is bad because they're often the results of deep persuasion, sometimes coercion, sometimes desperate decisions made under duress) but certain ways of experiencing and organizing my understanding that I chose long ago at the expense of other ways and other organizations, and that I've since forgotten I've chosen, or that those other ways are even possibilities. But then all of those choices were to preserve something and survive somehow and I suppose what I need to preserve and what needs to survive is 'me.'

Phenomenologically, my favorite writing on the topic actually comes from Denise Riley, a philosopher/poet whose essay "A voice without a mouth: inner speech" is indispensable reading for this stuff. She brings out, very nicely, the impersonality of the 'inner voice', which, I think if one really attends to it, becomes quite obvious. Some choice cites (sorry for the length but the essay is too good):

"As to how I inhabit my own inner speech, I am probably more accurately described as talking with myself rather than to myself. A great deal of what goes on in the head consists of an agitated running self-interrogation; ‘What did I just come into this room to get? Oh, I know, I probably left my cup of coffee in here, it’ll be cold by now. No, no, I had to telephone someone, but who was it?’ That is, there is an internal dialogue, but in these exchanges I appear to occupy both sides of it, and there is no one heavily weighted side to my garrulous split self.

...Still, perhaps at times a sense of dialogue can spring up in me, and I may feel that I’m talking both to and with myself when I notice that I have become my inner companion. Then I can silently calm myself, debate with myself. More censoriously, as my knowing superego I can berate myself, upbraid myself, goad myself along. But very often I do not actually address myself at all, and there is simply talking inside me. There is a voice. Questioned as to its origin, I would be in no doubt that it’s my own voice, but its habitual presence in me resembles a rapid low-grade commentary without authorship, rather than any Socratic exchange between several loquacious and attentive inner selves. Better Beckett’s accurate assertion: ‘whose voice, no-one’s’.

... We might say that inner speech itself lives as a state of ventriloquy, in that there is talking within us as if we are spoken from elsewhere; but this state just is our main mode of speaking. It’s present in the ordinary experience of overhearing myself speaking inwardly in a well-formed voice, whether as an outcome of switching my attention onto my inner speech or of feeling it to have risen and swum forward to claim my attention. Ventriloquy makes this daily inner speech: the state of sensing that words are running through me, across me. There is a kind of ‘it is speaking in me’ which is not exactly ‘it is speaking me’, but is an unwilled busiedness which I catch and may try to inhibit in myself.

Words race across me in polyphonic brigades, constantly swollen by the forces of more inrushing voices, and I can put up only a rear-guard censoring action. But this impression is no fully blown hallucination, for again there is no disowning and projecting of my inner voice, only my feeling of becoming a vehicle for words from elsewhere, much as a ventriloquist’s dummy or doll is made to speak vicariously. The real speaker’s, the ventriloquist’s voice, is thrown as if to issue from the passive doll, seemingly animating it. But the person who is the terrain of imperative inner speech, whether of love or hate or some other force, herself becomes the theatre for the performer and the puppet alike. The performer here is the arch-ventriloquist, language.

... Inner speech is no limpid stream of consciousness, crystalline from its uncontaminated source in Mind, but a sludgy thing, thickened with reiterated quotation, choked with the rubble of the overheard, the strenuously sifted and hoarded, the periodically dusted down then crammed with slogans and jingles, with mutterings of remembered accusations, irrepressible puns, insistent spirits of ancient exchanges, monotonous citation, the embarrassing detritus of advertising, archaic injunctions from hymns, and the pastel snatches of old song lyrics."

The choices, in my experience, warp around a core, and what surprises me as the years pass is how little what happens to me changes me, and how much I see in stages manifestations of the same old self in different clothing. Even traumatic things, I understand that I react to them in certain ways and they result not in a new person but reflect the one who was traumatized. Part of it is, having lived longer, I can now recognize myself more easily, and so I am becoming increasingly sympathetic to perennial rationalist truisms, and less impressed by their historical dissidents (empiricism, Buddhism, analyticism, continentalism), which seem by comparison like passing fads, and built on promises (of scientific advance, ending suffering, solving aporia) rather than being philosophies organic from life as lived. I want to see myself as I am, not as I wish I would be.

"As to how I inhabit my own inner speech, I am probably more accurately described as talking with myself rather than to myself. A great deal of what goes on in the head consists of an agitated running self-interrogation; ‘What did I just come into this room to get? Oh, I know, I probably left my cup of coffee in here, it’ll be cold by now. No, no, I had to telephone someone, but who was it?’ That is, there is an internal dialogue, but in these exchanges I appear to occupy both sides of it, and there is no one heavily weighted side to my garrulous split self. — StreetlightX

I don't know, I'm not schizophrenic. Maybe schizophrenic people think like this, but I do not. I do explicitly carry on dialogue with myself at times, but this is not the normal mode of my inner speech at all, but a dialectical tactic. I'm not Gollum.

Words race across me in polyphonic brigades, constantly swollen by the forces of more inrushing voices, and I can put up only a rear-guard censoring action. But this impression is no fully blown hallucination, for again there is no disowning and projecting of my inner voice, only my feeling of becoming a vehicle for words from elsewhere, much as a ventriloquist’s dummy or doll is made to speak vicariously. The real speaker’s, the ventriloquist’s voice, is thrown as if to issue from the passive doll, seemingly animating it. But the person who is the terrain of imperative inner speech, whether of love or hate or some other force, herself becomes the theatre for the performer and the puppet alike. The performer here is the arch-ventriloquist, language". — StreetlightX

This sounds like a description of a mental illness to me. I had a friend who was convinced he channeled demons, and it seems like something he would say.

Part of my interest in this is how much of philosophy, and philosophical pronouncement, is affectation. In my opinion the continental tradition is especially prone to affectation, whereas analytic philosophers are more prone to stifling.

You don't have to actually articulate thoughts as the inner voice as even preparing to say something is already enough to know where the articulation was headed, and to be responding to that accordingly.

It takes about half a second to assemble a fully fleshed out speech act. Hearing yourself say it as mental imagery certainly helps in provoking more precise responses in turn. You can stop to think about what you just suggested. Yet that is overkill for most trains of thought which are more the chaining of familiar habits. We know the thought was already going to be right and so no need to listen in with any care.

So inner speech is essential in that it gives human thought its rational structure. But then the aural image of a completed speech act is not essential. The latent structure can carry most of the load.