What Derrida Means When He Talks About Meaning

What the hell does this even mean???

“Meaning” is a multivalent term, which means that any attempt to discuss it meaningfully will be fraught with difficulty. Jacques Derrida makes a similar point when he asks, “Is it certain that to the word communication corresponds a concept that is unique, univocal, rigorously controllable, and transmittable: in a word, communicable?” Yet difficulty doesn’t entail impossibility. In the present paper, I wish to outline several distinctions that cut across the nebulous concept of “meaning” and which serve to alleviate some of its thorniness. First, we need to differentiate between (a) semantic meaning (often called “literal meaning”), (b) communicative meaning (which involves a speaker’s intentions) and (c) informative meaning (the non-semantic and non-communicative content expressed by an utterance). Second, we ought to distinguish between issues of epistemology (truth, certainty, etc.) and those of semiology proper.

In light of these distinctions, the Derridean account of meaning—insofar as it’s articulated in “Signature Event Context”—loses much of its initial unintelligibility. Specifically, it appears that the purported disjunction between meaning and a speaker’s intentions only holds with regards to semantic meaning. And even then, such a claim needs to be qualified. In what follows, I will first discuss the importance of separating semiology from epistemology. Next, I will delineate the notions of semantic, communicative and informative meaning. Finally, I will investigate the relationship between meaning and a speaker’s intentions.

1. Semiology vs Epistemology

In “Signature Event Content,” Derrida’s discussion of meaning is frequently punctuated by epistemic worries. Regarding the notion of interpretive context, he writes that “a context is never absolutely determinable, or rather, […] its determination can never be entirely certain or saturated.” And elsewhere, he criticizes J. L. Austin’s failure to acknowledge the possibility of misinterpretation. He writes, “Austin does not ponder the consequences issuing from the fact that a possibility—a possible risk—is always possible, and is in some sense a necessary possibility.”

Notwithstanding the importance of such qualms, we should be careful to bracket questions of semiotics (e.g., what is meaning?) from questions of epistemology (e.g., is it possible for us to reliably interpret the utterances of others?). First, these epistemological concerns seem to depend upon some prior conception of meaning. Indeed, an analogous concern motivates the ignostic, who argues that the concept of God must be defined before it can be accepted (theism), rejected (atheism) or even entertained (agnosticism).

Second, an utterance needn’t be “absolutely determinable” in order to be meaningful. A “one-time pad” code, for example, will be determinable only by those with access to the key. But it doesn’t follow that the ciphertext is thereby meaningless. More generally, unless it constitutes a categorical and epistemic impossibility that an utterance be correctly interpreted (an extremely bold claim), the analysis of meaning shouldn’t hinge upon any epistemological considerations.

Accordingly, my discussion of meaning will involve one modest assumption: that meaning may occasionally be transmitted from one person to another by means of language.

2. Semantic, Communicative and Informative Meaning

Semantic meaning is the meaning that an utterance has in virtue of certain linguistic conventions. Take, for instance, the sentence “Pass me the bat.” Such an utterance presents us with an enigma. We don’t know who uttered it, nor what—if anything—is denoted by the words “me” and “bat.” Nor is it clear whether we should read “bat” as “bat qua sports equipment” or “bat qua winged mammal.”

Nevertheless, “Pass me the bat” seems to be meaningful in a way that, say, “‡▷∎☆♬” is not. First off, the former phrase is ostensibly written in English, which already provides us with a set of guidelines or rules for interpretation. For instance, while “bat” could be read as either “bat qua sports equipment” or “bat qua winged mammal,” it will not mean “the Eiffel Tower.” Second, despite the sentence’s ambiguity, it seems possible to paraphrase “Pass me the bat” in a way doesn’t seem to affect its sense—e.g., as “Pass the bat to me” or as “The bat: pass it to me.” That which survives the translation from “Pass me the bat” to “Pass the bat to me” is the utterance’s semantic meaning.

Despite the intuitive import of these examples, semantic meaning remains a difficult concept to pin down by means of any rigorous definition. In this regard, I agree with Derrida’s statement that “the value of the notion of literal meaning [sens propre] appears more problematical than ever.” Yet it doesn’t follow that semantic meaning (or sens propre) is therefore an impenetrable concept. At the very least, the notion of semantic meaning depends on (a) an utterance’s material features (i.e., its phonemes, graphemes, etc.), as well as (b) some set of linguistic conventions.

Communicative meaning, by contrast, encompasses a series of related phenomena wherein an utterance suggests or implies something above and beyond its semantic sense. Austin invokes this notion when he draws a line between the locution and the illocution of a speech act. Consider a situation in which A says to B, “Do you know the time?” Presumably, the locutionary aspect of A’s utterance would be something akin to its semantic meaning. Now, let’s imagine that the exchange transpires as follows:

A: “Do you know the time?”
B: “Indeed, I do.”
A: “…”

Here, we would likely say that B missed the point of A’s query. For A was not merely curious about the extent of B’s knowledge (the literal import of A‘s question), but rather wished to be told the time. Therefore, the utterance “Do you know the time?”—as spoken by A—has a meaning apart from its literal or semantic sense.

The philosopher Paul Grice has also argued that most conversational exchanges make use of extra-semantic meanings. In “Logic and Conversation,” he provides the following hypothetical conversation:

A: “My car’s run out of gas.”
B: “There’s a gas station just around the corner.”

Surely, A‘s utterance is not functioning as a simple relation of fact. On the contrary, it likely has an illocutionary dimension, which we might paraphrase as “Help me, my car’s run out of gas.” Grice further explains that we tend to assume certain conversational maxims and principles—e.g., that our interlocutors are speaking with (some degree of) relevancy. Yet in order for B‘s response to A to be relevant to the conversation at hand, he would have to believe that the gas station was open, that it had gas to sell and so forth.

Accordingly, we may (roughly) paraphrase the communicative dimension of the conversation between A and B as follows:

A: “Please help me! My car’s run out of gas.”
B: “There’s a gas station around the corner, which should still be open. You
can buy some gas there.”

Here, both A and B are exchanging communicative meaning insofar as they are implicating or suggesting meaningful content beyond the mere semantics of their words.

As these examples show, communicative meaning involves the intentions of the speaker. It needn’t follow that the meaning of an utterance is infallibly “fixed” by a speaker’s intent. Telepathy notwithstanding, the intentions of others are as unassailable as the unobservable universe or the Lacanian Real. Still, a speaker’s intentions are that which determine whether or not an utterance has been interpreted correctly, in much the same way that a particular state-of-affairs determines the truth-value of a sentence like “The universe contains an even number of stars.” Intent is therefore an external condition of communicative meaning: an addressee doesn’t need to know that her interpretation is correct in order for it to be correct.

Moreover, for an utterance to have communicative significance, it must be at least possible for the addressee to work out the implicated or suggested content, which means that communicative meaning will form a two-part relation between a speaker and her intended audience. An utterance isn’t communicatively meaningful just because the speaker wishes it to be. Rather, it must somehow accord with the context of the exchange, as well as with certain conversational principles and maxims.

Thus far, I have discussed semantic meaning (which is the result of linguistic conventions) and communicative meaning (which is a function of a speaker’s intent). Yet as Derrida observes, “one characteristic of the semantic field of the word communication is that it designates nonsemantic movements as well.” Now, consider the following:

(P1) “The movie theatre ran out of pop.”
(P2) “The movie theater ran out of soda.”

Semantically speaking, these two sentences are identical. And we can surely imagine a scenario in which they don’t diverge much in their respective communicative meanings. Nevertheless, P1 and P2 express different information, such that the words “pop” and “soda,” as well as the spellings of “theatre” and “theater,” hint at the nationality of each speaker.

Other examples abound. An eloquently phrased utterance might bespeak the speaker’s learnedness, while garrulity could indicate her mendaciousness (i.e., “the lady doth protest too much”). Or an utterance may be interpreted according to non-linguistic conventions—e.g., the sentence “‡▷∎☆♬” might be read as {“diesis,” “right-pointing triangle,” … “double sixteenth notes”}. In each of these instances, what we’re looking at is an utterance’s informative meaning—i.e., that content which is not consciously transmitted by the speaker, nor fixed by any linguistic conventions.

The notions of semantic, communicative and informative meaning cover the various ways in which we use the word “meaning.” And although they haven’t been defined with any particular rigour (indeed, the very concept of meaning itself has been left unanalyzed), these notions will nevertheless allow us to investigate certain key issues regarding the nature of meaning, such as the vexed relationship between meaning and intent.

3. Speaker Intent

It seems that at least some interpretations require certain background beliefs. A semantic analysis of the sentence “Pass me the bat,” for instance, requires a prior knowledge of the English language—its definitions and its conventions. But what about the interpretation of communicative utterances? Let’s return to the exchange between A and B. In order to (correctly) interpret the sentence “My car’s run out of gas,” B must ascribe to A particular properties, beliefs and intentions—e.g., that doesn’t want to be out of gas, that she desires to change her present circumstances, etc. If this wasn’t the case, B wouldn’t be able to make the inferential leap from a descriptive utterance (“My car’s run out of gas”” to an imploring one (“Help me, my car’s run out of gas”).

Yet as Derrida observes, we can encounter the material features of an utterance (its phonemes, graphemes, etc.) without having any clue as to what the relevant background beliefs should be. Suppose that you come across small scrap of paper, on which you read the following enigmatic words:

(P3) “Expect warmly its tended garden”

Here, it seems as if we have an example of an utterance that has been completely divorced from its speaker’s intentions. What were the beliefs that gave rise to P3? Is the phrase “tended garden” meant to denote some actual garden? What is referenced by the possessive pronoun? Does the verb “expect” function as an imperative? And if so, is it directed at some particular addressee, or rather at any reader in general? Perhaps P3 is in fact part of some larger sentence. Perhaps it’s a fragment of a poetic couplet such as “I thought I might / Expect warmly its tended garden,” in which case P3 is not an imperative but (ostensibly) a description of a propositional attitude.

In other words, P3’s illocutionary aspect is indeterminate. Because the context surrounding its original utterance has been lost, we can’t reliably ascribe any properties, beliefs or intentions to its speaker. It follows that the sentence “Expect warmly its tended garden”—at least insofar as it’s presented here—lacks a communicative dimension.

And yet P3 still strikes us as somehow meaningful. Like “Pass me the bat,” P3 is (ostensibly) written in English, which provides us with a framework for interpreting its linguistic components. Moreover, the meaningfulness of these components can be demonstrated by the fact that we can use them to create new and meaningful sentences, such as “My mother keeps a tended garden” or “When I visit for the holidays, my family generally treats me warmly.” Derrida has written that “by virtue of its essential iterability, a written syntagma can always be detached from the chain in which it is inserted or given without causing it to lose all possibility of functioning, if not all possibility of ‘communicating,’ precisely.” Indeed, it appears that P3 is both meaningful and entirely divorced from its speaker.

Considering that an utterance can plainly “produce effects independently of [the speaker’s] presence and of the present actuality of his intentions [vouloir-dire]” (so says Derrida), should we conclude that meaning and intent are unrelated notions? No, we should not. First off, the iterability of utterances seems to be primarily a semantic property. Unlike the semantic sense of an utterance, its communicative and informative meanings depend upon conditions other than its material features or certain linguistic conventions. If we wish to iterate A‘s “My car’s run out of gas,” for instance, we’ll also need to repeat those conditions upon which its communicative meaning depends. These will include relevant background beliefs, as well as the circumstances of the original

Of course, we can use the material features of A’s utterance to say something else entirely. But even though the sentence “My car’s run out of gas” may indeed be (to quote Derrida) “detached from the chain in which it is inserted or given without causing it to lose all possibility of functioning,” its communicative meaning—insofar as it’s a function
of a certain linguistic and extra-linguistic context—would likely be lost. For while an utterance may be divorced from its speaker and still remain meaningful, this meaning will be semantic, not communicative.

It follows that the Derridian disjunction between meaning and intent applies most readily to semantic meaning. Even so, semantic meaning is not entirely free from matters of context. Take, for instance, the notion of homonymy. While the word “soy” qua a particular plant and “soy” qua the Spanish verb for “I am” bear identical spellings, it isn’t clear that they thereby constitute one and the same word. Granted, a Dadaist could dismantle a sentence like “Yo soy estudiante” in order to write, say, “I love soy milk.” But assuming that a word isn’t merely the aggregate of its graphemic and/or phonemic features, such a recontextualization would in fact change the very nature of the original word.

The grapheme “soy” thus expresses different words depending on its linguistic context. By way of analogy, a branch might be used as a walking stick, but it doesn’t necessarily follow that it is a walking stick. Comparably, while the word “soy” qua “I am” has the capacity or potentiality to express the English word “soy” (and vice versa), they do not thereby form the same word. The upshot is this: were we to stumble across a fragment of paper on which nothing more than “soy” was written, we’d have no reliable means for identifying which word had been inscribed—i.e., the semantic meaning of the utterance would be indeterminable.

To the extent that an utterance is essentially composed of its semantic, communicative and/or informative senses, it will not be perfectly iterable. Nevertheless, a semantically meaningful utterance will be iterable within a specific linguistic context (e.g., the context of English speakers), whereas a communicatively meaningful utterance may be iterable within a particular social group (as is the case with “inside jokes”).

One needn’t claim that the correct interpretation is always the best one. The Shakespearian exchange of wit, for instance, relies upon (intentional) misinterpretation. And many great intellectual developments have arisen out of the misconstrual of some earlier work. Besides, a misinterpretation may be pragmatically equivalent to a correct reading. Say, for instance, that B wishes for A to close the window. Puffing up his chest, he states in the most authoritative voice he can muster, “Close that window!” A, misconstruing B’s non-derogable command for a polite request, nonetheless obliges. Thus, although B‘s speech act has failed on one level (A didn’t take it to express a command), it succeeds on another (certainly, a general aim of B‘s utterance has been


In “Signature Event Context,” Derrida underscores an important aspect of meaning: namely, that utterances can remain meaningful even when divorced from the intentions of their speakers. The sentence fragment “Expect warmly its tended garden,” discussed earlier, is a prime example example of this. As it turns out, this sentence, despite its clear semantic significance, wasn’t produced by a conscious being at all. It originated via computational algorithm, with the complete “utterance” reading as follows:

In no impression assistance contrasted. Manners she wishing justice hastily new anxious. At discovery discourse departure objection we. Few extensive add delighted tolerably sincerity her. Law ought him least enjoy decay one quick court. Expect warmly its tended garden him esteem had remove off. Effects dearest staying now sixteen nor improve.

Yet to limit meaning to this—to pare it down to whatever can be conveyed by a machine and nothing more—strikes me as eliminativist in the extreme. Rather than trimming meaning down to the level of bare semantics (à la Derrida), we should instead embrace it in its various forms.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s