More than anything else, the trajectory of my career has been characterized by simple good luck – in particular, by serendipitous timing. I will not pretend otherwise. I had the very good fortune to begin my MA studies in Miami (where I now teach) when it was about to become the hotbed of the rule- and language-oriented version of Constructivism that I had been introduced to as an undergraduate, the version that, going forward, I will refer to as Constructivism simpliciter without any modifier. I had similarly good fortune in beginning my doctoral studies in Baltimore when it was a (albeit not the) US hotbed of the “Cambridge School” History of Concepts that I was then only superficially familiar with through its being invoked in works of IR Theory like Walker (1993), Bartelson (1995), and Onuf (1998). In neither case was I aware of the hiring that would be shaping things at these universities, so it did not influence my decision; it was truly happenstance.
Conversely, however, sometimes, (apparent) good luck yields bad results. One outcome of my involvement in the present project has been confronting again the signal error I made in an earlier publication (2003, 52). While I was surely lucky to have been involved in that earlier project at all, it was ill luck that my error was never caught prior to publication. This is not to deny my own responsibility for that error or my carelessness in making it. On the other hand, it would be remiss of me to fail to note the other mysterious failings pointed out in Michael Shapiro’s (2004) review. Failings that remain “mysterious” because they bear no relationship to any part of the subject matter of the essay or, indeed, any claim made in it, i.e. my purported failing of having simply rehearsed “now-familiar debates between a traditional neorealism, which still dominates IR, and the increasingly influential critical margin” in a piece I foolishly thought was a response to challenges for Speech Act Theory arising from developments in truth-conditional semantics in Analytic Philosophy of Language. Ultimately, however, I count myself lucky that my actual error – regarding one of the very fundamentals of Speech Act Theory – has never been pointed out in print, barring by one very merciful tenure reviewer, but that does nothing to diminish my embarrassment in having made it.1 Luck has its limits, and it should never be assumed to be working to your benefit.
In what follows, I will, in the next section, situate the two intellectual influences referred to above, Constructivism and Cambridge School History of Concepts, with regard to their common debt to Speech Act Theory, and highlight the differing purposes to which they put that approach. From there, I will turn to the tensions that trying to work within both traditions has brought about in my own ongoing work. In the final section, I will also address one of the questions motivating this volume: how I work; not in terms of methods, but rather a discussion of ethos and attitude, and how those factors affect me professionally.
The employment of Speech Act Theory – like any other theory – is a tool or tactic. As such, it can be put to multiple purposes. J.L. Austin (1955, 1961), by providing a neat delineation between the illocutionary purposes and perlocutionary effects of saying things, set the stage for an important divergence in how that theory might be employed. If one were to focus primarily on the illocutionary purpose behind utterances [I will, with some license, use “utterance” to refer both to verbal and written uses of language], Speech Act Theory can provide a useful set of interpretive resources. Conversely, if one were to focus primarily on the perlocutionary effects brought about (or that were intended to be brought about) by utterances, Speech Act Theory can provide a useful set of resources for analyzing social practices and institutions. As a shorthand, I will refer to the former use as “making meaning” and to the latter use as “making worlds”, but I do not want to make this distinction too categorical. The latter cannot happen without the former, and it is a key tenet of Constructivism that the former will necessarily result in the latter. I find the distinction useful because any author might legitimately be concerned with only one of these two dimensions (illocution and perlocution) at a given point. I use this language to try to highlight the different scopes of application to which historian Quentin Skinner and IR theorist Nicholas Onuf put Speech Act Theory (recognizing that Skinner rejected the label “theory” – although this says more about his conception of theory (“putting forward an hypothesis”) than about the content and substance of Speech Act Theory2 (Skinner 2002 [1988], 106).3 He nonetheless follows convention in referring to it as a theory.)
Both Skinner and Onuf employ Speech Act Theory, and Onuf makes use of Skinner’s historical work, although he does not touch upon Skinner’s use of Speech Act Theory (Onuf 1998, 65–7). This made me curious, but I think the explanation is simple: Skinner is concerned primarily with illocutionary force, what the speaker is doing in or by saying something, because he is interested in conceptual history, in understanding past usages of concepts and how those usages have changed. Onuf and constructivists in his tradition (among whom I number myself), by contrast, are concerned with both the illocutionary force and perlocutionary effects of speech acts, what is done or brought about by saying. This is because Constructivism is interested in social theory, particularly in explaining how our institutions came to be the way they are.
For Skinner, the relevance of Speech Act Theory lies in establishing authorial intentions; it is a tool for the interpretation of historical texts. Here, determining what the speaker is doing in saying what s/he says in a text is a necessary part of understanding that text; “whenever we use language for purposes of communication, we are always doing something as well as saying something” (Skinner 2002, 2). Just looking at what an author said (the locution) cannot give the full meaning of the utterance. In other words, it is not enough to know the meaning of what was said solely in terms of the sense and reference of the words used and sentences formed; we need to understand what they meant to do by saying it (Skinner 2002, 103–4). What was the author trying to do by making a given utterance (assertion) (Skinner 2008a, 650–1)? Without this information, we are, Skinner argues, missing “an entire dimension of understanding” (2002, 113).
The reception of speech acts and any subsequent perlocutionary effects, however, plays less of a role for Skinner. This is probably because his interest is limited by his textual subject matter mostly to assertive speech acts. Reception, inasmuch as it plays a role in his histories, is looked at primarily in terms of acceptance of the validity of the propositional content qua “prevailing conventions governing the treatment of the issues … with which the text is concerned.” (2002 [1972], 101–2). Skinner is not principally concerned in these analyses with speech acts as means of getting their recipients to act. He is fairly explicit about this in “Motives, Intentions, and Interpretations” (2002 [1972]) and “Conventions and the Understanding of Speech Acts” (1970), but he does give perlocutionary effects their due in his discussion of moral and ideological tracts in “Moral Principles and Social Change” (2002 [1974], 149).
Returning to his focus on the interpretation of texts, for Skinner, understanding “the illocutionary act performed by a speaker will be equivalent to understanding their primary intentions in issuing their utterance” (2002 [1972], 98). Knowledge of “a writer’s intentions [is]… equivalent to understanding the nature and range of illocutionary acts that the writer may have been performing in writing” (2002 [1972], 100). Figuring this out requires familiarity with the relevant conventions, whether semantic, syntactic or topical. Conformity with conventions is a prerequisite of the intelligibility of any speech act:
Familiarity with those conventions is a prerequisite of our later interpreting what the author may have meant to do by uttering those words. Speech acts may thus be thought of as a window into an author’s intentions. If we are to talk about speech acts at all, we have to make a few Wittgensteinian allowances regarding intention. Skinner’s starting point is the postulate that if someone has “engaged in an … act of communication, there must have been something that he was intentionally doing in saying what he said” (2002 [1988], 119). We know this because they chose to say what they said, and because they are trying to do something by saying it. “The illocutionary acts we perform are identified, like all voluntary acts, by our intentions” (2002 [1988], 109). Skinner continues, “the intentions with which [someone] is acting can be inferred from an understanding of the conventional significance of the act itself” (109).
This is where Wittgenstein comes in:
To assuage the concerns that Brent articulates about the problem of interiority in his chapter, (pages 191–192) let us look more closely at how Skinner has cashed out his understanding of intention. We are, as he says, “speaking of intentions embodied in acts of linguistic communication. But these intentions, including as they do the intention to communicate and be understood, will ex hypothesi be publicly legible in the manner of any other voluntary act” (Skinner 2008a, 150. Emphasis added.)
This is not the interiorist understanding of intention that Brent has carefully set aside, but neither does it fully fit his exteriorist criteria. In the end, it may come down to the question of whether the concept of intention is useful to our respective efforts. For my work, retaining such a chastened notion of intention – one that recognizes that intentions will not always be discernible – is a useful interpretive tool; for Brent’s project it obscures more than it clarifies, and allows too much when used as a means of moral exculpation – a sentiment I share, for example with regard to the Doctrine of Double Effect (Gould 2014).
So, both Constructivism and Cambridge School History of Concepts are in agreement that saying is one form of doing. Skinner is primarily interested in what has already been done (said). Analysis of a writer’s possible illocutionary purposes is a necessary element in understanding what has been said. As we have seen, with the exception of the hortatory writings of moralists and ideologists, however, he is not primarily concerned with the subsequent perlocutionary element, what the recipients do on account of the speech acts (2002 [1974], 145 ̶ 57) This element – creating states of affairs by saying, enacting by saying – carries the greatest amount of weight for constructivist applications of Speech Act Theory. We (constructivists) are interested in what is brought about; we are interested in how states of affairs are brought about; we are interested in bringing about, in making. Constructivism is – to that extent – poietic. These matters are less important for Skinner, but they are central for us. We are interested (to change idiom momentarily) in the relational aspect, upon what is attendant to assertive, commissive and directive speech acts. Skinner does not need to concern himself with the latter two categories, because the texts in which he is interested for the most part do not rely upon them in communicating to readers. We (constructivists), however, must concern ourselves with them because promises and commands are as important to any social interactions as the bare factual propositions of assertions, and, in many circumstances, they characterize social arrangements (institutions).
As we saw, rules in the guise of conventions play an important role in Skinner’s treatment, but it is a role that is both antecedent and exogenous to his approach. Rules work as limiting conditions on the effectiveness and intelligibility of speech acts for Skinner. He gives little attention to where these conventions come from, how they operate, or how they change. For his purposes, he may not need to. Skinner likewise gives secondary attention to the role of the texts he is interested in interpreting as potentially changing rules, including perhaps the very rules of intelligibility in question. In the works in question here, Skinner does not focus upon a link between speech acts and rules in any causal direction other than “rules [conventions] limit speech acts”. That speech acts might create or change rules does not seem to figure for him, except tangentially in e.g. “Moral Principles and Social Change”, where he says that speech acts can signal resistance to societal rules when using evaluative – descriptive terms. (2002 [1974], 149)
Since Skinner is not primarily concerned with these matters, he does attend primarily to the world making effects of all of this (Skinner 2002 [1972], 99).4 It is precisely those effects that interest Constructivism most. Even in the historical texts of Political Theory and Philosophy that Skinner works with, there are, of course, perlocutionary elements attendant to what these authors asserted. If readers accept the veracity of their statements of fact, if they accept the cogence of their arguments, this may, in principle, entail acting in ways that reflect that acceptance. It may, in other words, entail changing how they act. This is a change in the world. It is a change (or source of change) that Skinner turns to in his later discussion of paradiastole and its role in changing how we talk and how we attribute value to concepts and practices (Skinner 2007).
To explicate the constructivist usage of Speech Act Theory, I borrow in the coming paragraphs from my 2012 essay, “Cicero’s Ghost”.5 Assertive speech acts, the kind Skinner’s sources tend to focus upon, are articulated in the indicative mood, and have the following structure: X verb Y, where X denotes some noun, and Y may denote another noun or an adjective/condition. “X is Y;” “X counts as Y”. Such assertive speech acts make a claim about some part or parts of the (social) world. “Assertives … are speech acts stating a belief, coupled … to the speaker’s wish or intention that the hearer accept this belief” (Onuf 2013, 87). If the assertion is agreed to by the speaker’s interlocutor, this has the additional effect of thereby making that state of affairs be the case between them. This is a moment of social construction.
Saying that “X is Y” is – in some instances – what makes X into Y. The illocutionary purpose at work here is to get the interlocutor to regard or understand X as Y, or to get her to understand Y to be the case. The perlocutionary effect, should the speech act be successful, is to get the interlocutor to treat X as Y, that is, to act in such a way that s/he treats X in the way in which s/he treats Y, and hence with all of the attendant consequences of being Y.
The process does not end here. Subsequent to widespread acceptance of the assertion, a series of directive speech acts made in the imperative mood may follow. Their purpose is to “present the hearer with a speaker’s intentions as to some act the speaker would like to have performed” (Onuf 2013, 87).
Upon this scaffolding of assertive speech acts and directive speech acts, a third form, commissive speech acts, perform an important function. Speech acts of this form “reveal the speaker’s intention of being committed to a course of action” (Onuf 2013, 87). They may also follow assertives; indeed, they will often be intelligible only in light of the acceptance of the assertives introduced previously. Commissives will probably come in conjunction with the directive speech acts; in effect, they rearticulate all of the “you shalls” as “we wills”. They signal commitment to the factual assertions, and to acting upon the consequences of those factual assertions. “Directive and commissive speech acts are intended to have the same effect, which is to get the perlocutionary subject to do something” (Onuf 2013, 87).
Within the constructivist tradition, what follows from the successful reception of speech acts can be the creation of rules. These are quite easily derived from what we have so far seen; in the case of rules deriving from assertive speech acts, we get Instruction Rules stated in the indicative mood. In the case of rules derived from directive speech acts, we get Directive Rules, which like their precursor speech acts, are stated in the imperative mood. From commissive speech acts, we get Commitment Rules. These will be prominent in efforts at codification; they add normative weight to Directive Rules by signaling an express commitment to those rules; they likewise signal commitment to the initial assertion, recognition of the authoritative character of that assertion, and the authoritative character of the resulting Instruction Rules.
Skinner’s method shares much of our (Constructivism’s) understanding of language, and especially meaning, and what he uses it for is complementary to our aims. In particular, Skinner’s approach, with its emphasis on intention, retains an important place for authorial agency – one albeit delimited by structures of rules, and that is familiar terrain for us (Skinner 2008b). We are using the same tools for similar, but not identical aims. As such, we are using them somewhat differently. We use parts of Speech Act Theory that Skinner and those in his tradition do not because our aims require them, and theirs do not.
This is, I think, enough compare-and-contrast. I want now to turn to the problems I am facing while attempting to work with one foot in each tradition. As a constructivist, I am interested in how our current concepts (or institutions or practices, etc.) came to mean what they mean; for me, this is the heart of the constructivist project, but this is not a universally shared characterization. I am currently working on a monograph focused on the history of the concept(s) and practice(s) of prudence, analyzing its current denotation(s), and how they came about. Along the way, however, I am encountering a tension between the Cambridge School approach to the history of concepts and what I am finding in the arguments of the authors whose treatments of prudence, its antecedents, and cognates I am working with. Although not to the same extent as genealogy, this approach to history is keenly attentive to discontinuities and contestations over meaning, leading Skinner to famously say, “there is no history of the idea to be written”, but only histories of their “various uses and of the varying intentions with which it was used” (Skinner 2012 [1969], 85). Skinner famously views the works he is analyzing as contributions to dialogues, and hence as interventions in discrete, context-specific conversations or debates. (Skinner 2008a, 148) We should thus not treat the claims in these works “so much as statements of belief, but rather as interventions in the intellectual disputes of their time” (Skinner 2011, 275). We might equally think of them as moves in games or moves in arguments. Doing so, we need to ask why that move was made and try to recapture “the presuppositions and purposes that went into the making of it” (Skinner 2002 [1988], 115; 2008a, 148)
Turning this to the topic of prudence brings a problem to light: throughout the course of the two and a half millennia that this concept and its antecedents have been the subjects of scholarly discussion, authors have repeatedly insisted upon identity of meaning and denotation. For example, the Stoics held that their usage of phronesis was the same as Plato’s earlier understanding; in De Officiis (I.153), Cicero asserted that by prudentia he meant what the Greeks called phronesis (as if it had had only one meaning for them); Kant, following Christian Garve, rendered prudentia as Klugheit in discussing and translating Cicero, and understood himself to be speaking of the same thing as Cicero in his discussions of Klugheit in the Groundwork (4:397; 402; 442). Aquinas depended upon his treatment of prudentia in the Summa (II.II.47–51) and in the Quaestiones Disputatae (V, passim) being understood as Christianizing elaborations upon Aristotle and Cicero, even recapitulating the latter’s claim of conceptual identity with the Aristotle’s usage. There are many additional examples down to the present day, but the point is that there is a clear practice of authors insisting on conceptual continuity.6
I am confronted with the question of how to evaluate the claim of continuity when our method points toward these discussions and invocations of prudence being discrete moves in discrete games, and our job being “to grasp why it seemed appropriate to make just that move, and hence to issue just that utterance” (Skinner 2008a, 651. Emphasis added). Skinner acknowledges that “different pursuits have always been marked by the employment of relatively stable and characteristic vocabularies”, but does not address a situation exactly like this (2002 [1969], 58). If one of an author’s illocutionary acts (one of their moves) was saying that their term X means/is identical to what previous authors had meant by term Y, then we need to ask what it is that they are doing by making that assertion of identity. We need to address this question perhaps even before turning to their particular claims about the characteristics/role/prerequisites/value of the concept itself.
This question and its logical priority becomes all the more important when considering the very significant discontinuities in the substantive claims about prudence, and – as centrally concerns me in the manuscript – discontinuities in its attributed moral valence. What are we to make of the claim by one author that s/he means the same thing by a term as another author separated by centuries or millennia who then proceeded to make significantly opposing claims about its character? Kant is a signal case-in-point; his claims about Klugheit in the Groundwork, although identified by him with Cicero’s virtue prudentia, do not make it look like something to be cultivated at all. He held that he meant by his term what Cicero meant by his, but then radically devalued it, not treating it as a virtue to be cultivated at all, but as something inferior to morality.7
If we move beyond analyzing philosophical uses of concepts, however, it is easy to see how different users of the same word can claim that its referent is the same all the while attributing opposite valuations to it. Take, for example, “fish”; to my wife that word means a delicious food that is too rarely served in our home; to me it means a malodorous atrocity posing as food but which is fit only for our cats. But while we may both be able to point to some concrete-object-in-the-world and agree that it is fish, this does not necessarily scale to concepts or abstract nouns. The assertion “By fish, I mean what my wife means when she speaks of fish” does not work for me. It is not a shorthand for a bundle of denotations and valuations, and is mostly just a weird thing that no one ever needs to say, in part, because most of the objects of our lived world are not contested in the way that our concepts very often are.8
Deprived of that seemingly easy solution, the question of the Kant/Cicero example remains. A solution might lie in the earlier discussion of the role of convention in the intelligibility of speech acts. One possibility is the lingering role of convention requiring an argument to be linked to a prior authority’s claims; this works well to explain Aquinas’ invocations of Aristotle in his medieval, religious context. This is at the heart of Skinner’s employment of Speech Act Theory to recover the (intended) illocutionary force of utterances, and the reason that convention plays such a key role.
Another possibility is that the invocation is being made not as a signal of fidelity, but instead as a way to set up the older conceptualization for criticism or rebuttal. This would, however, either require assuming bad faith on the part of the author making the invocation, which is a fraught undertaking, or demand being able to show that the latter author distinguishes the meaning of the term as used by the original author from the valence of the term as used by the original author, but does not articulate that distinction.
As Brent has for several years always (and increasingly correctly) hastened to mention, I have been working on my prudence book for a very long time. This is a fair point, but I am dispositionally unwilling to put pen to paper until I am sure that there is nothing left to read on whatever topic I am working on. This – to my clear professional detriment – means muddling through texts in languages I am barely able to read; reading all of the footnotes of everything I read, and reading all of the things cited in them, then reading all of their footnotes, and everything cited in them. For me, this regress only reaches an endpoint when I no longer find anything unfamiliar being cited. I expend a lot of time is before ever writing by working in this fashion, sometimes to very little or no payoff. Respectful of my obligations as a professor, I am careful to use myself as an example for my students to urgently spurn if they ever want to get anything finished and graduate, but for me, the fear of my work being superficial is staggering, and that means never missing anything. And that means writing relatively little. My work is neither profound nor path breaking, and I do not pretend otherwise, but it is always meticulous; however, there is damn little of it.
Digging just deeply enough on a topic to say just enough about it is sometimes a necessary part of getting things written, submitted, and published – my own professors and my colleagues have always worked hard to make me see this. Accepting this advice is important, because it is a part of the business of academia. Churning out publications is valued over carefulness, and if one is uncomfortable in the sense Brent discussed, then s/he has no real choice in the matter but to work in that way. In my narcissistic and highly privileged view of the world, however, digging just deeply enough is a sign of carelessness and lack of seriousness as a scholar. This is why the standard academic model of ‘I am going to apply this theory to these three case studies’ is so tiresome. The engagement in that type of work with the theory in question in a book’s token ‘theory chapter’ or an article’s ‘theory section’ is usually just deep enough to allow the author apply it to some cases or to ‘test’ it via those cases, without ever really engaging the logic and evolution of the theory for its own sake. That very engagement, however, is, for me, why we have chosen this vocation. Obviously, not everyone wants to engage rather than employ or ‘test’; it is obviously necessary that the latter be done, but too much effort in that direction renders what we do a very dull sort of techne. Again, I am under no illusion about my work being intellectually deeper, more insightful, or really in any way better than any other mode of scholarship; I claim only that it is meticulous and exacting, and I worry that those two values, those two scholarly virtues are endangered by current Fordist norms and the “culture of speed” that they force upon us.9
Looking at these confessions through this volume’s framing, my way of work is obviously professionally self-defeating. These are not tactics for someone who aspires to ever be a widely recognized or high-impact scholar. At my current rate of work, I likely have at most two more monographs in me after my someday-completed prudence volume, not even putting me then at the point long since passed by many of my star contemporaries. To quote (anonymously) another tenure letter “[Gould] will never have a huge a citation count”. Neither will I ever be getting invitations to speak about my work at universities around the world nor have symposia held on it – I haven’t earned that kind of recognition, and am unlikely to. I am jealous but also resigned. I have learned to be content to write for myself in my way, and to recognize the freedom to do so for the privilege it is, a privilege that is increasingly denied in academia.
Adamson, Sylvia, Gavin Alexander and Katrin Ettenhuber (eds.) (2007) Renaissance Figures of Speech. Cambridge: Cambridge University Press.
Austin, J.L. (1975. [1955]). How to Do Things with Words, 2nd edn, edited by J.O. Urmson and Marina Sbisà. Cambridge, Mass.: Harvard University Press.
———(1961) Philosophical Papers, 3rd edn, edited by J.O. Urmson and G.J. Warnock. Oxford: Oxford University Press.
Bartelson, Jens (1995) A Genealogy of Sovereignty. Cambridge: Cambridge University Press.
Gentry, Caron E. and Amy E. Eckert (eds.) (2014) The Future of the Just War: New Critical Essays. Athens, Ga.: University of Georgia Press.
Gould, Harry D. (2012) “Cicero’s Ghost: Rethinking the Social Construction of Piracy” in Struett et al., Maritime Piracy and the Construction of Global Governance. Abingdon: Routledge.
———(2014) “Rethinking Intention and Double Effect” in C.E. Gentry and A.E. Eckert (eds.) The Future of the Just War: New Critical Essays. Athens, Ga.: University of Georgia Press.
Onuf, Nicholas G. (1998) The Republican Legacy in International Thought. Cambridge: Cambridge University Press.
———(2013 [1989]) World of Our Making: Rules in Rule in Social Theory and International Relations. Columbia, S.C.: University of South Carolina Press.
Shapiro, Michael (2004) Review of Debrix, ed. Language, Agency, and Politics in a Constructed World. Perspectives on Politics 2 (2): 423–424.
———(2002 [1988]) Visions of Politics: Volume I – Regarding Method. Cambridge: Cambridge University Press.
———(2007) “Paradiastole: Redescribing Vices as Virtues”, in Adamson et al., Renaissance Figures of Speech. Cambridge: Cambridge University Press.
———(2008a) Is It Still Possible to Interpret Texts? International Journal of Psychoanalysis 89 (3): 647–654.
———(2008b) “Interview with Quentin Skinner.” http://www.history.ac.uk/makinghistory/resources/interviews/Skinner_Quentin.html.
———(2011) “An Interview with Quentin Skinner. Contemporary Political Theory 10 (2): 273–285.
———(2012 [1969]) On Politics and History: A Discussion with Quentin Skinner. Journal of Intellectual History and Political Thought 1 (1): 9–31.
Struett, Michael J., Jon D. Carlson and Mark T. Nance (eds.) (2012) Maritime Piracy and the Construction of Global Governance. Abingdon: Routledge.
Walker, R.B.J. (1993) Inside/Outside: International Relations as Political Theory. Cambridge: Cambridge University Press.