Is Speech Act Theory Marxism?: Tim Keller, Pulpit & Pen, and Social Justice
Recent comments by Tim Keller have incited reactionary critiques aimed at the concept of speech act theory. Critics claim that speech act theory, as employed by Keller, is a Marxist concept that makes a speaker responsible for every unintended effect of speech, greasing the wheels for conceiving of speech as violence. Here, we will explore what speech act theory is, its limits, whether it is a Marxist concept, what Keller could have meant by using it, and some advice for future dialogue surrounding this issue.
1. What Speech Act Theory Is
Speech act theory is an approach to language which resists the reduction of linguistic acts to their propositional content. It attempts, rather, to account for everything which a communicator could possibly do that counts as communication. All of these possibilities, in this theory, fall into three categories of communicative action—locutionary acts, illocutionary acts, and perlocutionary acts. Each of these acts can occur simultaneously in a single communicative act.
A locutionary act is simply an utterance—it is the sum total of semantic meaning, phonetic and rhetic aspects—phonetic being the audible component of speech, and rhetic the informational. An illocutionary act is what is done with a word. There are multiple things that can be done with words, including assertives (expressing a proposition), directives (requests and commands), commissives (promises), expressives (“congratulations,” “thank you”), and declarations (baptisms, marriage declarations). A perlocutionary act is the attempt to elicit some response in a recipient.
For example, if my wife said to me over dinner, “I don’t like your cooking,” there are locutionary, illocutionary, and perlocutionary elements to this act. The locutionary element is composed of the sounds she uses, the rules of the English language she conscripts, and the information about her culinary preferences. If she said, “No me gusta tu cocina,” that would have been a different locutionary act, even though it portrayed the same information. The illocutionary element is assertive—that is, she is making a claim. She didn’t say, “Cook better next time.” That would be a different illocution. That would be a command. By saying, “I don’t like your cooking,” she is asserting something.
Her perlocution is ambiguous. What in me did she intend to elicit by saying, “I don’t like your cooking?” Does she hope that by asserting a claim about the poor quality of my cooking, I will improve? Or is she simply trying to hurt my feelings? Determining the perlocution, from the listener’s perspective, is usually much harder to discern than the locution and illocution.
Speech act theory was popularized by philosopher J. L. Austin, and later John Searle. Austin distinguishes between a perlocutionary objective, which is the intended outcome of an illocution, and a perlocutionary sequel. The perlocutionary objective is the intended effect, and the perlocutionary sequel is the actuated effect. Nevertheless, “A judge should be able to decide, by hearing what was said, what locutionary and illocutionary acts were performed, but not what perlocutionary acts were achieved.”
Austin is ambiguous about whether perlocutions extend to unintended consequences, but Searle seems to say strictly that perlocutions do not extend to unintended consequences—rather, perlocutions are merely intentions. This seems more appropriate, since an act is performed by an agent, and insofar as perlocutions are conceived as acts, their boundaries ought to be the extent to which agents can act, which is bounded by the limit of an agent’s actionable power.
2. The limitations of Speech Act Theory
Vern Poythress outlines several limitations of speech act theory that make its analytical utility dubious. He points out three issues which impede the linguistic cash value of the theory as the issue of hierarchy, the issue of variation, and the issue of distribution. These are three ways that the theory can run into trouble.
First, issues of hierarchy. Searle says that “The characteristic grammatical form of the illocutionary act is the complete sentence (it can be a one-word sentence).” But speech act theory, by honing in on one behavioreme, one sentence, one paragraph, or even one document, can fail to see layers of locutionary, illocutionary, and perlocutionary acts which go up and down the entire spectrum of specificity. For example, a politician may be performing an illocutive request, which may be part of an illocutive apology, which may be part of an illocutive political speech that is essentially expressive or assertive. Each of these illocutionary layers, which can happen simultaneously, serve as locutions for the greater illocution. So, speech act theory, as a tool for linguistic analysis, has as its unit of analysis not only the atomic aspect of language, but the metalocutive aspect of language, or its greatest and most cumulative unit of semantic measurement.
Second, there is an issue of variation. What this means is that, to use Poythress’s example, a politician might ask, “Please do not judge me harshly.” In classifying this utterance as a request, we permit its reduction to that illocution, and nullify the illocutionary nuance of the whole locution. So, we are happy to say, “In other words, he is requesting mercy.” But the phrase is doing so much more than that, that this summary would be untrue due to its reduction.
The utterance “Please do not judge me harshly” both avoids any admission of guilt and makes the request politely, indicating a reverence for the audience and social etiquette. Therefore, by calling the utterance “Please do not judge me harshly” an illocutive request, we don’t actually add anything meaningful to our understanding of that utterance in particular, but rather help us to relate that utterance to, and distinguish it from, other utterances. And so, speech act theory, when it comes to the variation of the linguistic elements for which it seeks to account, has as its object language as a whole, rather than particles of speech, even though it insists the opposite.
Third, there is the issue of distribution—that is, the complexity of a statement. So a single utterance may be loaded with an assertion and a command–“You are charged with (assertive) … I hereby sentence you (declarative).”
Fourth, and Poythress makes this point informally—there is the issue of being in a position to perform a speech act. For example, while a declaration of marriage is a speech act, you must have credentials to marry recognized by the state. You can only call a ball or a strike if you are an umpire—and more than that, if you are a home plate umpire. You can only apologize if you are the one who perpetrated the moral wrong for which you are apologizing.
Those are four qualifications that we need to make about speech act theory before we place it on the front of our hermeneutical toolbelt. The problem of hierarchy, variation, distribution, and position.
3. Speech Act Theory and Theology in Recent Discourse: A Case Study
“There’s a thing called speech act theory. Speech act theory … says ‘You can’t just analyze words by what they say. You have to analyze words by what they do.’ … You could say, ‘I love the way you look.’ … Under certain circumstances, that could be a rather … coercive statement. … In other words, there’s what it says, and then what is it trying to do? … In the end, what concerns me most about it is not so much what it’s saying, but what it’s trying to do.
And, what I fear is, and I think they may be responding to a set of sins on the other side — at this point, I feel that the Christian church is being, because our country is so polarized politically, that increasingly, the church is increasingly becoming an extension of the various political parties. …
If you talk about the evangelical church, for example, there’s now becoming a red evangelicalism and a blue evangelicalism. It’s almost like there are … churches that are lining up and becoming more extensions of a particular political party than they are really looking at what the word of God says.
Because, I actually do think that … there’s a lot of things the Bible says about sex and gender that really sound out sounding politically (today) conservative. And a lot of things the Bible says about race, and justice, and the poor, that (today) come out sounding extremely liberal.
And therefore … the church cannot identify so completely with one party. And I think that’s what this thing ends up [doing]. It’s not so much what it says. It’s what it does. It’s trying to marginalize people who are trying to talk about race and justice. It’s trying to say, “You’re really not biblical.” And, it’s not fair in that sense.
If somebody starts to go down it with me, and say ‘Would you agree with this? Would you agree with this?’ I would say, ‘You’re looking at the level of what it says, and not at the level of what it’s doing.’ And I do think that what it’s trying to do is that it’s trying to say: ‘Don’t make this emphasis. Don’t worry about the poor. Don’t care about the injustice. It’s not really that important.’ That’s what it’s saying. Even if I could agree with most of it, I don’t like it. … It’s what it’s doing that I don’t like.”
Pulpit & Pen published an analysis that said this:
“To summarize Keller another way, the words contained within the Dallas Statement may be true, but they are still bad because of their ramifications. Unable or unwilling to explain any theological deficiencies in the statement, Keller cites Marxist philosophical tenet called Speech Act Theory.
This tenet is a subset of the discipline of pragmatics, and it holds that words not only convey information (their literal reading) but they can also be used to carry out actions. Speech-Act Theory was developed by American philosopher, J.R. Searle. Searle is a professor at the University of California, Berkeley, and he is a protege and fan of Jürgen Habermas, the German Marxist and supporter of the Neo-Marxian Frankfurt School. Speech Act Theory has been pivotal in the aggression of Marxism in the fields of both philosophy and linguistics and has been essential in developing the notion of verbal microaggressions. Through Speech-Act Theory, Marxists have popularized the notion that words can be bad, even if they’re true if they cause adverse side-effects. Seemingly innocent and innocuous speech can be sinister if it’s a subtle cue from the majority to further oppress the victim class.”
4. Reasons for the Misapplication of Speech Act Theory
There is a lot to unpack here, but this theological powder keg can be diffused with a few simple remarks.
First, speech act theory is not a Marxist concept. It does not, as a concept, entail that the author of a speech act is responsible for all unforeseen harm caused by his speech. In the terms we delineated earlier, speech act theory does not entail an ethical scope in which a speaker’s moral domain extends beyond his intended perlocutionary objective to every unintended perlocutionary sequel and sequelae.
Second, Searle has explicitly denounced association with Habermas, particularly on this very point of Marxism. The above quote at Pulpit & Pen betrays a lack of understanding about how the Frankfurt School differs from revolutionary Marxism at a conceptual level. But Searle comments: “I’ve never been able to take the ‘Marxism’ in the Frankfurt school very seriously. By ‘the Frankfurt School,’ I mean the contemporary Frankfurt school, primarily Habermas and Apel.” So here, Searle distances himself from that school. But more basically, the reason that Searle distances himself from this approach is a matter of genre. While Habermas was a continental philosopher intent on making meaningful philosophical claims about human life, Searle was interested in tracing the very grammar of human speech. Searle, as an analytic philosopher, was looking for something like an abstracted mathematical formula that best explained the possibilities of human speech. In this sense, Searle’s project is not only not Marxist, but resists the sort of synthetic philosophical project that would find economic analysis helpful at all. The summary of Pulpit & Pen represents a Googled knowledge of these issues.
Third, Keller’s conscription of speech act theory borders on misappropriation. He could be doing one of three things in utilizing speech act theory here—he could be saying that he agrees with the locution of the social justice document, but classifies the rhetic content not as asserting propositional content, but as declaring the theological unimportance of the politics of monetary impoverishment, and possibly even the unimportance of ethno-specific suffering. Second, Keller could be claiming that the authors of the document had an intended perlocutionary objective to place a white perspective at the center of the conversation about social justice, and to marginalize ethnic minorities. Third, Keller could have meant that he found distasteful the perlocutionary sequel of the document, and that he could agree with the locutionary and illocutionary aspects of the document.
To read Keller charitably, I think we could say that this third option is most likely the case, since the first two would make Keller guilty of the logical fallacy of mind reading against the authors of the social justice statement. But we should make two points here.
First, Keller’s Westminster Philadelphia background makes him sensitive to aggressive theological politicking that always has someone in its crosshairs. The Philadelphia presbyteries of the PCA and OPC, with whom Keller had fellowship during the time he was closely associated with Westminster Theological Seminary, are so full of aggression and militancy that Keller is probably used to sniffing out nefarious and hyper-vigilant attacks behind every theological statement of the sort in question. This doesn’t excuse a misattribution of nefarious intent to the authors of the statement, but if Keller’s primary point is to attribute malintent to the statement’s authors, and he is in fact guilty of disagreeing with the illocutionary or perlocutionary object of the statement, then I can understand autobiographically why Keller would have this knee-jerk reaction, even if it is not accurate.
Second, if Keller dislikes the social justice statement merely because of its perlocutionary effect, then he doesn’t need speech act theory to do this. He can just say, “I don’t like the way it’s been received. I don’t like the associations that have stuck to it. Therefore, I don’t like it.” He doesn’t need speech act theory to distinguish between a document and its effects. As an ordained Presbyterian minister, Keller subscribes to the Westminster Confession of Faith, even though it was used to marginalize religious minorities in England, and was even used by Robert Lewis Dabney to justify slaveholding. The point is that perlocutionary effect has nothing to do with the document itself, and it remains somewhat unclear with what and why Keller takes issue with this document from a speech act perspective, which is exactly what he uses to explain his position.
In summary, while Keller’s take on this statement and his use of speech act theory are muddy, the rejection of speech act theory on this basis is not only absurd, but flies in the face of commonsense truths established by 20th century analytic philosophers on which Keller’s respondents clearly don’t have an operating grip.
5. Guidelines to Avoid Misapplication of Speech Act Theory
We can see that the profundity of speech act theory lies in its giving us a way to highlight meaning that transcends the propositional, or the rhetic, aspect of any given locution. It gives us a way to classify the intent of a speaker when they are not merely descriptive. Beyond that, it is not profound. For example, we don’t need speech act theory highlight an author’s deeper intent, an author’s desired outcome, or the subtext of a conversation. All speech act theory does is put these tools in a single taxonomy for classifying utterances, but the same rules of interpretation apply to our understanding of those utterances—our interpretations of them must be evidence-based, and not over-attribute meaning where it does not exist. Speech act theory may seem to warrant the psychoanalysis of the speaker, licensing a paranoia about nefarious action beneath the speaker’s intent without evidentiary warrant. This would be a misappropriation of the speech act theory tool, and this violates the way language works to serve dialogue between two parties.
I’ve heard Progressive Christians use speech act theory both as a reason for rejecting inerrancy and as a way to make speakers responsible for the effects of their speech, rather than their intent. They use it as a way of wiggling out of propositional truth commitments, such as confessional statements, and of overattributing nefarious or irrational intent that doesn’t exist. Neither of these uses of speech act theory are appropriate to the theory itself. They are bastardizations of the concept, and use it as a way of sophisticating and minimizing logical fallacies crucial to their arguments.
Speech act theory is not a model for mapping the ethical entailments of speech, but is a taxonomy of the actionable possibilities available within the tool of language. Put simply, speech-act theory is just a list of things you can do by talking.
While Keller doesn’t really utilize the concept of speech act theory correctly — whether he was intending to highlight the social justice statement’s intent (which is dubious) or effect (which is more objective) — his point about the statement’s reception can still offer something valuable. Keller is a pastor in New York City, and those in his circles likely didn’t receive the statement well. Keller also came out of the Westminster Theological Seminary context in Philadelphia, which is a hotbed of theological divisiveness and politicking in the PCA. Therefore, due to Keller’s past experience with hyper-militant ecclesiology, and his present context in a very leftist city, it makes sense why he would perceive the social justice statement the way he does.
Does he use the concept of speech act theory incorrectly? Kind of. Does he conflate the document’s illocutionary class with its perlocutionary sequel? In this video, it seems as if he does. So, this allows us to extrapolate out three elements from this conflict which we ought to retain — the original claims of the Statement, the cultural reception of the Statement, and the utility of speech act theory. Keller’s video muddles them all together, but watch-bloggers basically accept that muddling and throw the philosophical baby out with the cultural bathwater. We need speech act theory to understand how language works. Perlocutionary objectives are real in the minds of speakers, but the moral responsibility of the speaker stops at the extent of this objective. We must distinguish between the perlocutionary objective, which is the intended effect of a locution, and the perlocutionary sequel, which may misattribute authorial intent that did not exist, as well as attribute positional authority over how a locution is received that no author has.
The criticism of speech act theory as Marxist is not informed, and perpetuates the sort of political dialogue that is willing to dispense with valuable conceptual material for the sake of getting a lauded jeer at one’s opponent. My advice to Keller would be to lay off the speech act theory for a few days and explain your ideas without an appeal to some esoteric philosophical work. If you actually say what you mean, rather than couching it in an obscure theory, critics and fans alike will actually have a better grasp on what you’re saying.
My advice to all critics of the theory, from someone very skeptical of the SJW and intersectionality movements: either go to seminary, or you’ll get taken to seminary. This level of conceptual dialogue is plebeian and perpetuates the sort of ambiguity that tolerates equivocation. My only remaining advice: Before you become a theological blogger, get a seminary degree or you’ll get taken to seminary by people who actually know what they’re talking about. It’s because of blogs like these that Mark Noll is able to credibly write what he does about the evangelical mind.
 Rhetic acts are acts of uttering phemes (words or sentences) “with a certain ‘meaning’ in the favourite philosophical sense of the word, i.e., with a certain sense of reference.” John Austin, How to Do Things with Words, 94. See Steward Thau, “The Distinction Between Rhetic and Illocutionary Acts,” Analysis 32, no. 6 (1972): 177-183.
 John Dore, “Holophrases, Speech Acts and Language Universals,” Journal of Child Language 2 (1975):
 John L. Austin, How to Do Things with Words, 2nd ed. (Cambridge, MA: Harvard University Press, 1975; orig., 19672); and John R. Searle, Speech Acts: An Essay in the Philosophy of Language (London: Cambridge University Press, 1969). See also John R. Searle, Expression and Meaning: Studies in the Theory of Speech Acts (Cambridge: Cambridge University Press, 1979); Intentionality: An Essay in the Philosophy of Mind (Cambridge/New York: Cambridge University Press, 1983); William P. Alston, Philosophy of Language (Englewood Cliffs, NJ: Prentice-Hall, 1964). Speech Act Theory is sourced from the later writings of Ludwig Wittgenstein, under the influence of Jürgen Habermas.
 "Saying something will often, or even normally, produce certain consequential effects upon the feelings, thoughts, or actions of the audience, or of the speaker, or of other persons: and it may be done with the design, intention, or purpose of producing them." John L. Austin, How to Do Things with Words, 101.
 Austin, How to Do Things with Words, 122.
 John R. Searle, Speech Acts: An Essay in the Philosophy of Language (New York: Cambridge University Press, 1969), 46.
 Vern Poythress, “Canon and Speech Act: Limitations in Speech-Act Theory, with Implications for a Putative Theory of Canonical Speech Acts,” Westminster Theological Journal 70 (2008): 337-354.
 Searle, Speech Acts, 25.
 Searle admits as much: “Often, we do more than one of these [speech acts] at once in the same utterance.” Searle, Expression and Meaning, 3-4.
 Bruce Krajewaki, “Interview with John Searle,” Iowa Journal of Literary Studies 8, no. 1 (1987): 75 [71-83].