This is a claim at once absurd and viable I wish to develop a little further. I should say in advance that the development is speculative, metaphorical, under-theorised. I've been in too good a mood recently to have any decent thoughts. And perhaps the pull of the attractor basins of more workable understandings in our common conceptual currency will be too great to allow for the requisite patience for such ill-formed thoughts to be entertained by any sensible reader.
It is tempting to develop the miracle claim by comparing expressive with descriptive language. By, that is, describing the contrast in such a way as the miraculous character of the expressive can be compared with the banal character of the descriptive. But in the sense in which it is miraculous, descriptive language is no less expressive than expressive language. It is the 'expressivity' of all language which is miraculous. Which is, in a way, not to draw a real contrast, and so to risk not saying anything, or to risk mystical Tractarian posturing about showing instead of saying something. I mention these concerns because, well, because they have occurred to me along the way, presented themselves as more viable opportunities for the development of the putative thought than they were.
It is tempting to think that language, spoken or written, has a function. Of course we use this or that bit of language for a thousand functions. But these functions are defined internally to the uses of language; they are not determined from outside. So, yes, it is tempting to think that language has some overall function. To 'represent reality' perhaps. To 'convey information'. To 'convey ideas'. To 'describe'.
But in the relevant sense, language tout court has no purpose. Or better, it makes no sense to ascribe an overall purpose to language use. Our life with language is the field in which purposes are opened up as possibilities; what sense then to ascribe a purpose to discourse per se?
But then in what sense is language 'miraculous'? I think it seems miraculous when, at least, we approach it from an already crumbling instrumentalist perspective. When we had imagined, in our phantasy, that we somehow typically stand behind, stand back from, our language and put it to use. And then we realise that it can't be like that. I don't mean to say that this is the only perspective from which it seems miraculous; I shall return to this below.
Heidegger wrote that language speaks us, we do not speak it. (See Charles Taylor on 'Heidegger, Language, Ecology'.) In saying this he was speaking nonsense in order to remove a narcissistic prejudice. For whilst of course language does not literally 'speak us' (that doesn't mean anything!), and whilst of course we speak a language or a sentence and not language per se, to the extent that (as it were) it is (which it is not) true (rather than just meaningless) to say that we don't speak language, then to this extent it is true to say that language speaks us. But a better way to put it would be to acknowledge that 'we' and 'language' are internally related - whilst remembering that the former (if we take 'we' in a royal sense) has been around for a good deal less time than the latter. (Hence Heidegger's phrasology, or so I'm supposing.)
We start life by grunting, and gradually this grunting richens itself into discourse. Wittgenstein's developmental parable (Investigations, para 244) about the development of first-person sensation language works not just by countering an 'inner description' myth with a story about expression or avowal. Rather it works by reminding us of the ongoing immediacy of the relation, or better the non-relation, of the grunt to the sensation.
How do words refer to sensations? -- There doesn't seem to be any problem here; don't we talk about sensations every day, and give them names? But how is the connection between the name and the thing named set up? This question is the same as: how does a human being learn the meaning of the names of sensations? -- of the word "pain" for example. Here is one possibility: words are connected with the primitive, the natural, expressions of the sensation and used in their place. A child has hurt himself and he cries; and then adults talk to him and teach him exclamations and, later, sentences. They teach the child new pain -- behavior.
We do not (typically) feel a desire to grunt and then grunt; we do not find we have something to grunt about and then grunt. We just: grunt. Grunting is where we are at that moment. We are grunting. And so too with discourse. (cf Schreber's bellowing miracle. Perhaps what made Schreber's bellowing seem miraculous to him was that it refused to fit the dominant psychotic phantasy that he was instrumentally reesponsible for his own expression.)
"It is miraculous that we 'express our thoughts' in language." "It is miraculous that we 'understand' each other's utterances." This understanding is no less immediate (except when we don't at first understand - of course!) than the grunting. We are "there with each other" in language. We are born into it, we live it, it lives us. From the perspective of narcissistic instrumentalism, understanding comes to look like sense-making, sense-finding, interpretation, decoding. But that is what is 'so miraculous': this (apart from in all the interesting exceptions themselves proving, relying on, the rule) just doesn't happen. Understanding is there. You speak, I carry on. 'A packet of crisps please'. 'That will be 40 pence'. There: understanding, speaking; no thinking. This just does happen.
Or, if there had been thinking, then the very occurrence of that thinking - without prior thinking - would be - is - 'miraculous' in just the same way.
If it is almost sensible to speak of 'miraculous', this is because there is one very good sense in which it is impossible, senseless, to try to explain how we speak or how we understand. And it seems particularly miraculous because it comes all too naturally to take up a perspective from which it seems that some such explanation really ought to be forthcoming here. That perspective is the one from which we are, as it were, 'behind' our language, the instrumental narcissistic homuncular user of our language.
It is senseless, in this one very good sense, to ask or answer how we speak or understand, when that question is asking for an explanation of how thought or intention gets put into language, or how it gets read off from it. Attempting that quite general explanation once again quite gets the cart before the horse or, more absurdly yet justly put, gets the horse before itself. Augustine imagined that learning a first language was like learning a second language (Investigations para 1). The thing is, though, that in the relevant sense in which we might be inclined to say how we learn our first language: There isn't a how. Again - there are a thousand hows for a thousand specific questions, questions that arise within the life of the language user, but to think about those parts here is not to accept the dubious global invitation that philosophy is holding out for us.
A change of tack.
"Neurosis results from attempts to speak (or not speak) speech, to perform (or not perform) actions, to feel (or not feel) feelings." That is a bold claim, but I believe it is just. It helps explain, amongst many other things, why neurosis never 'works'. We try to speak our lives, but our symptoms just start to speak us. Neurosis is a fear, a hopeless flight from, the miraculous.
ACT, amongst other new-wave CBTs, asks us to accept our thoughts as just that, and to 'defuse' from them. (In fact it (rather incoherently) seems to go one step further, and views thoughts as private verbal events. But clearly a thought is not that, for we can have private verbal events which are not thoughts but which are, say, our inner soliloquising rehearsal of our lines for a play.) It encourages us to 'comprehensively distance ourselves' (defuse) from our thoughts, to stop treating them as necessarily reflecting the way things are, and to see them for what they are - i.e.: thoughts.
It occurs to me that, by encouraging a 'noticing' relation to own's own thoughts, we might run the risk of encouraging neurosis. That kind of self-alienation - in which one starts to exist in a cognitive or quasi-perceptual relation to one's own thought - in which the idea that one is one's thoughts - is just what neurosis amounts to.
So is there another way to describe what ACT is doing? Well, I think there is. It is - building acceptance of the occurrence of these thoughts, accepting that they occur, that they are moments of us - removing the instrumentalist control agendas that we neurotically develop in relation to our thoughts and feelings.
The relevant de-fusion is between thought and world, not between thought and self. Through this defusion, the thoughts are (as Richard Wollheim would say) 'brought under the regulative concept of' 'belief'; they are owned as such. And belief is a cooler place to stand than in the heat generated by the knowledge and delusion driven by desire.