Before I Speak: Derrida, Foucault, and the Problem of Beginning

1. Before I Speak

“I have but one language—yet that language is not mine.” (Derrida, Monolingualism of the other)

Derrida’s sentence sounds like a contradiction. To have a language normally means to speak it, to use it, perhaps even to regard it as part of who one is. Yet Derrida immediately withdraws the possession his words seem to claim. He has one language, but that language is not his.

Foucault begins from a different image:

“I should have preferred to become aware that a nameless voice was already speaking long before me…” (Foucault, The order of discourse)

Here the disturbance concerns not possession but beginning. Foucault imagines a voice that was already speaking before he arrived, so that he would not have to occupy the position of the one who starts.

The two sentences do not make the same argument. But they disturb a common assumption: that when I speak, I begin with myself. I imagine myself as the speaker who has something to say, finds the words for it, and sets speech in motion. Derrida makes the language itself precede the speaker. Foucault imagines discourse already underway before the speaker enters it.

The difference between these formulations matters less, for the moment, than the question they make possible. What if speaking is never quite the beginning we imagine it to be?

We normally experience words as ours. We choose them, arrange them, and use them to say what we mean. Yet the words were there before us. The language in which we formulate a thought has a history that does not begin with us. Even the simplest act of saying something involves entering a world of words already in place.

Before I can say something in my own words, there are already words.

That may seem obvious. But it becomes less obvious once we ask what this means for the relation between thought and language.

2. The Thought Before the Word

We often imagine thought as preceding its linguistic expression. First there is something in the mind: an idea, an intention, an experience, a judgment. Then we look for words that will convey it to someone else.

The model is simple:

I have a thought → I put it into words → I communicate it.

Everyday speech contains countless versions of this assumption. “I know what I mean; I just can't find the right words.” The sentence seems perfectly natural. It suggests that the meaning is already there, intact, while language is the means through which we try to give it adequate form. The words may succeed or fail, but the thought itself appears to exist before them.

Much of the familiar philosophical picture of expression allows us to imagine thought in this way: language comes after thought and gives it an external form.

There is nothing obviously wrong with this model. We do sometimes struggle to express something we already seem to understand. We can revise a sentence because it does not quite say what we intended, or translate an idea from one language into another. Such experiences make it tempting to regard language as a vehicle for content that exists independently of it.

But the model becomes less secure when we ask where the distinctions within the thought itself came from. The words we use do not merely label ready-made ideas. They provide distinctions, contrasts, and possibilities through which something can become articulate in the first place. Before I can decide that one formulation expresses my thought better than another, I am already working within a language that gives me those formulations.

The relation between thought and expression, then, may be less straightforward than the familiar sequence suggests. Language may not simply arrive at the end of a process that has already taken place in the mind.

What if language is not simply what comes after thought? What if the language in which thought becomes articulate is already involved in its formation?

That question changes the status of the words we call our own. If the language through which I think was already there before me, in what sense can I call it mine?

Derrida’s sentence gives that question an unusually precise form.

3. The Language That Is Not Mine

“I have but one language—yet that language is not mine.”

For the present argument, the force of the sentence lies less in its paradox than in its temporal structure. Derrida speaks a language that he did not create. He enters it rather than establishing it.

I do not invent the words with which I first learn to speak. I do not establish their grammatical relations or decide which distinctions the language makes available. By the time I begin to use it, a linguistic world is already in progress.

This is what gives the phrase “not mine” its force. The language is intimate without being self-created. It is the medium through which I remember, think, describe myself, and address others, yet none of these uses makes me its origin.

Entering a language, however, does not reduce the speaker to a passive user. We can bend familiar expressions, give old words new uses, combine them in unexpected ways, or say something that has never been said before. A language can precede us without dictating every possible utterance.

This is the important distinction: precedence is not determination.

The language comes before me, but what I say within it can still be mine. Indeed, the possibility of saying something new depends on entering a medium that I did not create.

This complicates the familiar picture of thought and expression. If language were merely a transparent instrument, the fact that it existed before me would be little more than a historical curiosity. But if language is already involved in making thought articulate, then its priority is more consequential. I do not first possess a fully formed thought and then select words from outside it. I become capable of articulating what I think within a linguistic world that I have inherited.

I inherit the medium in which I can say something that has never been said before.

The problem, then, shifts from the language we inherit to the discourses in which we enter. What happens when the words we use are already part of something that was underway before we arrived?

4. A Nameless Voice Was Already Speaking

Foucault’s opening of The Order of Discourse gives that problem a different form. He imagines that he would have preferred to discover that “a nameless voice was already speaking long before me.” The fantasy is not of remaining silent. It is of being relieved from having to occupy the position of the one who starts.

That position is more demanding than it first appears. To start a discourse is to appear as its source: I am the one from whom these words proceed. Foucault imagines another possibility. The speaking would already be underway; he would merely enter it.

The voice is significant precisely because it is nameless. Foucault does not imagine discovering that another identifiable person had spoken first. If that were the case, the problem would remain a simple question of priority: someone spoke, and then I spoke. But an unnamed voice points to something less personal. What precedes the speaker is not necessarily another individual. It is discourse already in motion.

This extends the problem raised by Derrida. Derrida asks how a language can be mine when I inherited it. Foucault asks what it means to speak when the discourse itself has a history that precedes the speaker.

The question therefore shifts. Instead of asking only what the speaker wants to say, we can ask: Who is speaking when I speak?

The question does not yet require us to identify the rules or institutions that organize discourse. The image comes first. Someone begins to speak, but the beginning is already strangely occupied. The sentence has barely started, and yet something is there before it.

Beginning to speak and being the beginning of speech are not the same thing.

5. We Never Speak from Nowhere

The point can now be stated more precisely: We do not begin from nowhere.

When I speak, I enter something already in progress. The forms available to me have a history; the distinctions I can make have not been invented for the occasion. Yet entering an existing medium does not mean merely repeating what was already there. I can rearrange it, redirect it, or use it against itself.

Two distinctions follow.

Beginning a speech is not the same as beginning the language in which it is spoken.

Saying something new is not the same as creating the conditions under which it becomes sayable.

These distinctions separate novelty from absolute origin. A statement may be new without emerging from nowhere. A speaker may interrupt an established conversation without having created the language, concepts, or forms that make the interruption intelligible.

This gives Foucault’s nameless voice its sharper significance. The speaker does not have to disappear for discourse to have a history. I can be responsible for what I say without being the first source of everything that makes my saying possible.

What comes before us does not close the space of what we can say. It is the space from which we begin. We enter an existing field, and precisely because we enter it, we can alter its arrangements.

The speaker has a beginning. Discourse does not begin there.

6. Before the First Word

We can now return to the two sentences with which we began.

Derrida writes:

“I have but one language—yet that language is not mine.”

Foucault imagines:

“a nameless voice was already speaking long before me.”

Read together, they unsettle a familiar picture of speaking. We often imagine a simple sequence: first there is a self, then a thought, and finally the words that express it. The two sentences suggest a more complicated order. Before the self can fully articulate what it thinks, it has already entered a language and a discourse that did not begin with it.

The speaker is therefore neither outside language nor simply erased by it. I can begin an utterance without being the absolute beginning of what makes that utterance possible. I can say something that has never been said before without having created the medium in which it can be understood.

Speaking begins somewhere. But it does not begin with the speaker.

Before I can say “I” in my own words, there is already a language in which “I” can be said.

References

Derrida, J. (1998). Monolingualism of the other, or, The prosthesis of origin (P. Mensah, Trans.). Stanford University Press.

Foucault, M. (1972). The archaeology of knowledge and the discourse on language (A. M. Sheridan Smith, Trans.). Pantheon Books.  

Foucault, M. (1981). The order of discourse (I. McLeod, Trans.). In R. Young (Ed.), Untying the text: A post-structuralist reader (pp. 48–78). Routledge & Kegan Paul.