What’s “Love” Got to Do with It?

Let me frame a sympathetic objection to an earlier post: “For the sake of argument, let’s suppose I accept your claim that tautaulogies (NP1 Be NP1, as in Love is love) are semantic as well as syntactic, rule-like as well as word-like. It’s plausible enough that they are part of the ideology of language, in the way that that the legacy of structuralism in critical theory taught us to see with words. And maybe humanists, to the extent that they’re invested in meaning and form, could start paying attention to the history, meaning, and ideological function of linguistic forms like it. I understand why, from your perspective, what has been subject to literary and cultural study in the last few decades looks not like the full domain of language, but only a relatively narrow band or stratum of language – only lexically specified sign units.

But even if you are right that there is more to see and explain about language than we find in our current lexicons, it doesn’t follow that humanists ought to train their critical sights on such things. Even if politics is pervasive in our language and does real work beyond our vocabulary, it doesn’t follow that scholars should be compelled to identify, analyze, and contests it workings everywhere, in all the strata of language.

The fault lines, the factional and social differences, that matter to our contemporary politics are – no surprise here – organized around differences of race, gender, class, sexual, identity, ability, and so on.   We might give somewhat different accounts of what categories belong on this list (caste, species, religion, national origin, etc.), or what words best name those categories. But no matter: they will be words. And the same is true if we study law, or desire and affect (“love”), or our political institutions, or poetics, empire, animals, climate change, disability… or discourse on nearly any topic. When we are interested in culture and history, the kinds of categories and concepts we will be interested in will reside in the lexicon, in the words we use to describe people, identities, institutions, and ideas, and not in tautaulogies and other abstract signs that operate like combinatory grammatical rules. In fact, when we talk about studying a “topic” or “subject,” what we mean (on at least on some accounts) is studying what is picked out by a small subset of our lexicon, by some words and not others.

So while cultural studies could, as your book argues and as your earlier posts suggest, re-gear itself with digital tools to study a wider range of signs, and might have a more ‘complete’ account of language as a result, doing so would be to take a step away from precisely those issues that matter or reasonably should matter most to the work of cultural studies.”


Ok, I think this is a fair point. (Why thank you, imagined objector!) The relative abstraction of tautaulogies and other lexically unfilled sign units like them – the very property that makes them the basis for linguistic creativity – also makes them less germane to the particular content-oriented questions that scholars, working in the cultural studies mode, set out to ask. To give an example, it is understandable that historians of human rights have not taken an interest in the hierarchy of nominal compounds that the term instantiates. And this will be true on a sliding scale: the more abstract the sign, the less lexical and conceptual specificity it will involve.

My response at this stage is speculative rather than self-assured. But I think there are good reasons why abstract signs like NP1 Be NP1 matter nonetheless, and why leaving them unstudied has left a big gap not just in cultural studies’ account of language but in its analytical and critical toolkit.

The formality of linguistic forms are not a distraction from our current political situation. In fact, the abstraction of a sign like NP1 Be NP1 – the constrained indifference of its variables – gets right to a (the) central feature of liberalism as a doctrine and a political regime.

Under this regime, the law assumes, or wishes to assume, a formal existence, and specifically an indifference (within its particular scope) to particular cases, particular people or classes of people, and their particular substantive moral doctrines and aims. (Here I set aside, for the moment, the rise of illiberalism under Trump, which has called this basic indifference into question.) Judges are supposed – or suppose themselves – to apply law in uncontroversial, and even mechanical ways, constrained only by the law itself and the facts of the case, and setting aside normative issues that would compromise the law’s autonomy. Except for a few narrow cases (the named classes protected from employment discrimination, for example) the law is written without respect to the identities of those who are subject to it. As the arch-classical liberal F.A. Hayek writes, “It is the Rule of Law, in the sense of the rule of formal law, the absence of legal privileges of particular people designated by authority, which safeguards… equality before the law” (117).

As Hayek was willing to acknowledge more directly than most of his successors, “formal equality before the law” in this conception “produces economic inequality” and is “incompatible” with the aim of “material and substantive equality.” What’s more, the legitimacy of liberalism does not actually require “formal equality before the law,” only the appearance of such equality. Hayek argues that what can be claimed for the Rule of Law is just that it is not “designed to affect particular people in a particular way.” But even this is claiming too much. All that law’s legitimacy requires under liberalism is that this design is implicit and plausibly deniable.

Some of the best-documented examples of the disparity between the formal indifference of law and the particularity of its history, design, application, and effect involve the war on drugs. In 2010, four years before Washington D.C. legalized the recreational use of marijuana, black people were 8 times more likely to be arrested than white people, despite using and selling pot at similar rates. Nationwide, black Americans arrested for drug charges were almost 6 times more likely to go to prison. These disparities persist despite the fact that laws regarding drug possession are “color blind,” with no mention of racial identity, and the politicians who made the laws had learned to expunge racial vocabularies from their public statements. But journalists and historians have brought to light compelling evidence that the unequal effects of the war on drugs were hardly incidental to its design. Legal scholars and sociologists have long studied the gap between the formal indifference of the law and the inequitable particularity of its application and effect.

As long as literary and cultural studies in the structuralist tradition works only with words as the sole bearers of meaning, the sole building blocks of utterances, and the sole protagonists of linguistic history, it will be unable to take account of the gap between rule and application, between the relatively abstract and law-like non-specificity of langue (language as a system), and the particularity of parole (what is said).

What I am proposing here is a rough analogy between the formal equality of law and the formal abstraction of forms like NP1 Be NP1. The NP “blanks” or variables in a tautaulogy are lexically unspecified, allowing them to be filled in ways that speakers haven’t encountered before, even as they are constrained to be noun phrases rather than other constituent constructions. But the blanks in tautaulogical constructions are never simply blank, never neutral or empty containers that can be filled equally by any NP. They have an internal structure that is the product of the form’s history. What can be said is built out of and configured by what has been said. This is what linguists mean when they call a theory of grammar “usage based” and what Derrida took account of in Glas when he wrote that langue is constituted by the contamination of parole. It is easier to repeat what has been said before than to invent anew – easier to repeat prefabricated instances such as Boys will be boys or girls will be girls than to produce unwitnessed instances, and easier, until quite recently, to say War is war than Love is love. Repeated use wears grooves into the blanks in our constructions, so that there are more and less obvious, more and less costly, ways to fill them.

(The name we give to those who break from the grooves worn into the blanks of our constructions is “poet.”)

Linguists have developed a range of ways of describing the structure of lexically unfilled blanks that are not simply a matter of frequency. Instead they make full use of our figurative capabilities – our ability to establish likeness and difference. Since George Lakoff adapted the prototype theory of the psychologist Eleanor Rosch to linguistic categorization, linguists have observed that many open blanks exhibit a “radial” structure, which means that they are organized around a central exemplar or exemplars, with less central members derived by metaphorical extension. Without further evidence from a corpus or archive, it’s not plain that the blanks of tautaulogies have a radial structure, but examples that do are not hard to come by. When we read I shop therefore I am or I eat, therefore I am, we know that shop and eat are not simply members of the category “verb” but also analogical replacements for Descartes’ exemplar think.

Understanding tautaulogies and other abstract signs requires resisting two kinds of reduction. On the one hand, they are lawlike elements of (langue) insofar as, in their abstraction, they are different from and irreducible to the totality of what is said (parole). A blank’s internal structure arises from the history of its use, but by virtue of abstraction it is never identical to that history. Conversely, a form never achieves the pure or autonomous formality of the categories Chomskyan rationalists suppose we possess at birth, before linguistic acculturation begins. Its blanks never pull free from the history of saying out of which they emerge.

Let me draw out more explicitly the analogy between linguistic forms and the form of law under liberalism. Both emerge from a concrete history, a history of lives lived and things said. Both necessarily acquire a degree of autonomy or “formal existence” that differentiates them from the concrete history that birthed them. It is this formality that allows a law to determine unforseen cases and allows a form like NP1 Be NP1 to license unwitnessed instances. In their formality, both are to one degree or another removed from the lexical specificity of the identity terms (black, white, male, female, gay, straight, etc.) that have preoccupied recent decades of literary and cultural theory in the Lacanian/Althusserian tradition. But their formality nonetheless arises from and is configured by a history of dramatically unequal treatment. They bear that inequality in the structure of their categories and reproduce it in their subsequent use – “colorblind” laws when they are applied to particular cases, lexically unspecified constructions when they are instantiated in our utterances. (It goes without saying that the formality of linguistic forms and the formality of law, while both matter, have quite different consequences for our collective life.)

A cultural theory unable to look beyond our inherited vocabularies is in the position of a legal studies capable only of critiquing those laws that make explicit mention of particular identities. Without understanding the categories of linguistic forms like NP1 Be NP1, we won’t understand how inequality is built into the structure of our language or the mechanisms by which that inequality reproduces itself.  At the same time, these forms are fundamental to our everyday creativity and our ability to remake that structure. More on that anon.


“Cultural Marxism” According to Stuart Hall c. 1983

Last post on Stuart Hall – at least for now. In his 1983 lecture series Hall traces what is at once an intellectual history of the Centre for Contemporary Cultural Studies at Birmingham, a (provisional) history of the development of cultural studies as a discipline, and a history of his own intellectual tradition.


Hall’s lineage runs like this: F. R. Leavis(!), Richard Hoggart, Raymond Williams (in conversation with E.P. Thompson), Claude Lévi-Strauss (drawing from Ferdinand de Saussure and Émile Durkheim), Louis Althusser (building on Lacan’s account of subject formation), Antonio Gramsci (with a side glance at Nicos Poulantzas), and perhaps (by extension, and certainly for us, now) Hall himself. Marx and Engle provide the ground base over which this tradition develops. There’s a page or so on Foucault in the midst of a longer treatment of Althusser’s account of the subject, and a glancing reference to Walter Benjamin. That’s basically it.




The most obvious thing to notice about this tradition is that it is entirely male. Hall was not, in 1983, unaware of women writers and intellectuals – he had been married to the feminist historian Catherine Hall since 1964 – or of feminist theory.  At one point he suggests, for example, that “the question of reproduction [of the social relations of production] has been assigned to the Marxist (male) pole, and the question of subjectivity has been assigned to the psychoanalytic (feminist) pole” (135). But at this stage of his career he did not regard women or feminist thought as a determining influence on his own intellectual formation.  (For an illuminating account of history of Stuart Hall’s initial resistance and eventual recognition of feminism, and of its place at the Centre, see Escosteguy).


Reading Hall and the intellectual tradition he traces got me thinking about “cultural marxism,” a term that has had a somewhat eventful history on the right over the last forty years or so, and that has recently returned to prominence in the conspiracy theories of the alt-Right and the memos of Trump officials.


As a term, “cultural marxism” is actually quite apt for Hall and the intellectual tradition he assembles. The culmination of that tradition is an analysis of the full domain of culture in terms of domination, hegemony, and struggle. At various points over the last thirty years, the term has on occasion been used to refer to Hall, Hoggart, and the Birmingham school more generally. Yet when the right warns of cultural marxism today, this is not the tradition or school they are talking about. Why not?
I’m afraid the obvious answer is also the right one: not enough Jews.


In its dominant version, the right-wing genealogy locates the origin of cultural marxism at the Institute for Social Research in Frankfurt, Germany. In this genealogy, members of the Frankfurt School – Jewish academics like Theodor Adorno, Max Horkheimer, and Herbert Marcuse – fled to the US with the rise of Hitler, bringing with them their analysis of mass culture and their plot to dethrone white, Christian, capitalist, colonial Western culture.


It’s rather surprising to me that “cultural marxism” has come to be so closely associated with these particular thinkers, this specific genealogy. In graduate school, I spent quite a bit of time reading and discussing Adorno, Habermas, and other members of the Frankfurt School, partly because many of my friends and interlocutors were in intellectual history and political theory working on the history of German philosophy, phenomenology, and existentialism.


But even as I was studying Frankfurt school thinkers, it was evident that its members were far from the dominant strain in the study and theory of culture. For Adorno, at least, some of this had to do with the arduousness of his prose.


But the bigger reason, I think, is that he and other members of the Frankfurt school never had an encounter with structuralist linguistics.  Unlike left theorists in France, they continued to speak in the idiom of German idealism and Hegelian dialectics. They lacked a holistic account of signification, and thus an account of how culture functioned as a system of signs.  Structuralism was inherently well-suited to the study of literary texts and other linguistic artifacts, which made for easy entry into literature departments.  As I examined in a previous post, Hall uses the analytical toolkit of structuralism to reveal the contingencies of our racial vocabularies, and along with many others he worked to adapt it to the full range of popular media (photography, film, TV, clothing, etc.) as well.   It’s this structuralist legacy that is of particular interest to me.  The aim of my new book, Cyberformalism, is to rebuild the structuralist toolkit – and especially its account of the linguistic sign – so that it can account for language’s combinatory capacities as well as its meaningful elements.


While members of the Frankfurt school are certainly an important part of the intellectual history of the new left and of the present moment, their lack of the structuralist toolkit as well as their reliance on an earlier philosophical idiom limited their influence – on the study of culture, in college classrooms, and in left political movements.


When the right tells a monogenetic story of cultural Marxism that leads back to a cabal of German Jews, they are doing bad intellectual history.  (It’s hard to credit anything said by Jordan Peterson, but at least he diverges from the ersatz story by accusing Derrida of infecting  “humanities types.”)  Much of this can doubtless be chalked up to unabashed anti-Semitism. But I suppose that the need for a simple story matters too – simple in the sense of originating in a single school, in the plot of an ethnically homogeneous people… who just happen to be Jewish.  It’s much easier than acknowledging, as Jeet Heer observes, that “Anti-racism, feminism, and the gay rights movement all have roots that well precede the Frankfurt School and owe far more to the activism of women, people of color, and LGBT individuals than to any German theorist.”  For the alt-right, it has the added feature of denying women, people of color, and LGBT individuals the capacity to pursue their own political interests without serving as the unwitting puppets of a group of Jewish men who have been dead for half a century.


Hall’s assembled intellectual history, despite being all-male, lacks the simplicity and homogeneity of the right-wing’s monogenetic tradition.  It includes Oxford professors born into the British working class; a traditionalist literary critic; French linguists, sociologists, and anthropologists; an Algerian-born French Marxist; an imprisoned Italian Marxist journalist; and, in Hall, a Jamaican-born cultural theorist of African, British, Portuguese Jewish, and Indian descent.  It’s not clear that those on the right who blame the current troubles of the US on the influence of cultural marxism are capable of taking account of this sort of heterogeneity.  But it is obvious enough why they don’t have any interest in doing so.


Stuart Hall and the Anonymity of Syntax

Lately I have been developing an account of how literary and cultural studies, along with intellectual history and related disciplines in the humanities, became lexicalist disciplines. That is, how did they come to study and understand language as what Reinhart Koselleck variously calls a “dictionary” or a “sociopolitical vocabulary”? How did questions about meaning and signification get narrowed to questions about the relationship between word and world?


One needs to take only a small step away from the lexicon to see many of the methodological developments of the last few decades – the study of keywords, Begriffsgeschichte, historical semantics, cultural semantics, critical semantics, literary semantics, queer philology, the architecture of concepts, and many of the methods that now fall under the umbrella of digital humanities and distant reading – as variations on a single, coherent project: historicist lexical semantics. Martin Jay has recently offered a useful survey of how this project, flanked by metaphorological and rhetorical counterparts, has changed and evolved in intellectual history over the last century.


I always hasten to add: I think the project of historicist lexical semantics is legitimate, productive, illuminating, and worthwhile. As an account of language, meaning, and creativity, however, it is also incomplete, limited, and limiting. My new book, Cyberformalism, tries to suggest how humanists might move past the lexicalism of historicist semantics to study other kinds linguistic sign units. But now I am asking: hud did the humanities acquire its lexicalist limits in the first place.


One text to which I’ve turned is Stuart Hall’s history of Cultural Studies, as he related it in a series of lectures delivered in 1983. There he gives an account of the structure of structuralism as, “an elegantly simple model consisting of elements and rules of selection and combination” (63).


The elements – words – get further attention, but the rules of selection and combination quickly retreat from view. Here is a passage where we can see that retreat as it happens.


In Jamaica, Where I spent my youth and adolescence, I was constantly hailed as “coloured.” The way that term was articulated with other terms in the syntaxes of race and ethnicity was such as to produce the meaning, in effect: “not black.” The “blacks” were the rest – the vast majority of the people, the ordinary folk. To be “coloured” was to belong to the “mixed” ranks of the brown middle class, a cut above the rest – in aspiration if not in reality. My family attached great weight to these finely graded classificatory distinctions and, because of what it signified in terms of distinctions of class, status, race, and colour, insisted on the inscription. Indeed, they clung to it through thick and thin, like the ultimate ideological lifeline it was. You can imagine how mortified they were to discover that when I came to England I was hailed as “coloured” by the natives there precisely because, as far as they could see, I was “black” for all practical purposes! The same term, in short, carried quite different connotations because it operated within different “systems of differences and equivalences.” It is the position within the different signifying chains which “means,” not the literal, fixed correspondence between an isolated term and some denotated position in the colour spectrum. (147)


This is a powerful example of how the structuralist account of language became an implement in the arsenal of critique. Tipping his hat to Althusser with the word “hailed,” Hall shows that the same word (signifier) has different meanings, marks out different identities, when located in a different cultural and linguistic systems. Two terms that are essentially synonyms in British English (coloured=black) instead mark an opposition or distinction in Jamaican English (coloured=not black). Hall’s analysis (which follows the trajectory of his autobiography) challenges the necessity of racial regimes by showing their contingency across languages and cultures (in other passages, across historical periods as well). Though this kind of analysis owes much to Hall, it does not belong to him or even to the category of race, but instead has roots and ramifications across all identity categories and all domains of cultural theory and critique.

Notice Hall’s use of the term “syntaxes.” What does it mean to speak here of “syntaxes of race and ethnicity”? One might suppose that the term refers to the “rules of selection and combination” that, along with meaningful elements, he earlier described as twin components of linguistic structure. But Hall’s analysis, however powerful, performs no analysis, and offers no examples, of “combination” – no sentences, syntagms, chains, or sequences of words – at all. There are only individual terms – “coloured” and “black”- and the semantic associations between them. (Perhaps we can suppose that they are members of a common paradigm, from which they can be “selected” and combined into any number of utterances). The upshot, in any event, is that “syntax” refers only to the “systems of differences and equivalences” between meaningful elements, between words. In Hall’s analysis – as in the structuralist tradition of which he is, I think, highly representative – “syntax” is fully absorbed into the lexicon.


There is nothing wrong with using the term “syntax” in this way – to indicate the relations between the words in the lexicon. Linguists who, since Chomsky, have focused on the structure of sentences have no monopoly on that term (or, for that matter, on the term “grammar,” which also occasionally gets used in the same way). But using “syntax” in the way that Hall does leaves anonymous and unaddressed the issue of creative combination. Without a name to call its own, at least in the structuralist tradition, the issue of combination – how a finite repertoire of meaningful elements are joined to make an indefinite number of complex utterances, and indeed to invent new elements – steps out of the frame.


Searching for Love is Love

In a previous post I tried to show why scholars of literature, culture, and intellectual history might want to direct their attention beyond the words in the lexicon to more complex and abstract signs. Specifically, I argued that tautological forms (tautaulogies, following Joyce) are conventional, Saussurean signs that do consequential political work in the domain of what Gramsci called “common sense.”

The post drew on a range of examples, such as Love is love, It is what it is, and The law is the law. But it didn’t attempt anything like the kind of longue durée history that is performed in the chapters of Cyberformalism (which, did I mention, is out now with JHUP?). The meaning and use of the form varies across languages, cultures, and history. How can we study that variation? How would we begin to piece together a history of tautaulogies, just as we now have abundant histories of words?

These questions are why there is a cyber- in Cyberformalism – why the book falls under the umbrella of the digital humanities. To study linguistic forms, you need to be able to find them. And for that task print finding tools are unsatisfactory. An index, card catalog, concordance, or dictionary will be of little or no use in locating “equative” tautaulogies such as Love is love and Poems are poems or when dogs were dogs, which share no word forms in common. Both include inflections of the verb be, but that fact, on its own, does almost nothing to winnow them from a text archive of any size.

Let’s think a bit about how digital methods might allow us to search for Love is love and other equative tautaulogies. I notated their general form as NP1 Be NP1. How closely our search criteria approximates that form depends on the capabilities of our finding tools. As we will see, search can get technical and complex pretty quickly, even running right up against the current limits of Natural Language Processing. But we can make some progress nonetheless.


Let’s say that all you have is a plain text archive and the ability to search it using regular expressions. You can test out which REGEXES match text strings with a tool like regex101.com.

Let’s start with (\w*) (were|was|is|be|are) (\w*). This string says, in effect,find strings of three words, where the middle word will be a form of Be.”   It matches love is love, but it also false positives such as dogs are mammals. And there are many other tautaulogies like Boys will be boys or The law is the law that it won’t capture.

In some interfaces, regex has a capability called backreferencing (or backref for short), which allows it to match the same text that it has already captured. (In language, as in cognition generally, the notion of “the same again” goes very deep, but backref offers a simple, mechanical version of it.)  So the string (\w*) (were|was|is|be|are) \1 means “find a word, followed by a form of be, and then the same as the first word again.” It will match love is love and Let Bartlett be Bartlett, but not dogs are mammals. Adding an optional will, as in the string (\w*) (will)? (were|was|is|be|are) \1, also captures Boys will be boys. But backref has limitations. Regex’s matching of letters is entirely, well, literal. If the text reads Love is love, the backreference will be sensitive to the case of Love and won’t match the lowercase love. If there is a workaround available, I would be glad to know it. Otherwise, more powerful tools are needed.

What about if we are searching in Part of Speech (POS) tagged corpora like those at CQPweb or the BYU corpora? Then additional capacities are available to us. We can search not just for words in general but for categories of words. In CQPweb the search string _N* _VB* _N* reads “any noun followed by any form of be followed by any noun.” If backreferencing were available, you could search for (_N*) _VB* \1, which would match War is war, love is love, girls are girls, etc.   And there are additional levels of flexibility. Instead of capturing just a single noun, you could capture an optional article as well with ((_AT*)? _N*) _VB* \1. While I don’t have a way of testing this out, it should match instances like A deal is a deal, and The law is the law, as well as Love is love. It’s also worth pointing out that a search doesn’t have to capture every kind of equative tautaulogy in a single search string. It will often be easier to run a number of distinct searches, each of which retrieves a subset of true positives.

REGEX and POS tags are never going to be fully up to the task of identifying abstract signs in text archives. In the terms of Chomsky’s Hierarchy, a Type 3 (or “regular”) grammar will always lack the power to fully generate (or match) human languages, which are generated by a Type 2 grammar (with occasional Type 1, or “context sensitive,” exceptions). But more powerful, Type-2-ish search tools are available.

Annotated by a full constituency parser, like the one from Stanford (http://nlp.stanford.edu:8080/parser/), Love is love would look like this:

(ROOT   (S     (NP (NN Love))     (VP (VBZ is)       (NP (NN love)))))

And a tautaulogy with a more complex Noun Phrase, such as The sanctity of law is the sanctity of law, would be annotated like this:

(ROOT   (S     (NP       (NP (DT The) (NN sanctity))       (PP (IN of)         (NP (DT the) (NN law))))     (VP (VBZ is)       (NP         (NP (DT the) (NN sanctity))         (PP (IN of)           (NP (DT the) (NN law)))))))

These embedded, hierarchical, labeled sentences are rich but unwieldy digital objects. There are, at present, few parsed-text archives of a size or accuracy that would be useful for cultural or historicist inquiry. Manipulating or searching them adequately would involve using a search language like Tregex or TGrep2, probably in conjunction with a scripting language like Python. This falls well beyond my technical capabilities. I would need to partner with a good computational or corpus linguist to search parsed texts for tautaulogies or other similarly abstract forms.

Humanities inquiry can sometimes appear to proceed as if it were a-technical, without material or instrumental support. But that is an illusion; as our tools grow familiar and get taken for granted, they become invisible. The supports of lexical research are all ubiquitous, present in virtually every kind of reference work, every book with an alphabetical index. One task of scholars of media and book history (as an early modern scholar, I am an especial fan of work by Ann Blair) is to remind us of the role that such reference technologies play in the production of knowledge.  Another is to show us that the reference technologies that seem second nature to us now (like alphabetization) were never obvious or easy when they were first being built and used.

It should be no surprise that studying new objects of philological, linguistic, cultural, and historical knowledge will involve grappling with research tools that are difficult, technical, complicated, unwieldy, and only partially adequate.  Formulating a search is not extrinsic to the forms that are searched for; it involves thinking through the very nature of those forms. Our conceptions of language develop in tandem with the material supports for that conception.


When America Was America: Cultural Critique Beyond the Lexicon


In a previous post I began to share some of my ongoing explorations into how the humanities (perhaps more accurately: the interpretive disciplines, especially those that study textual artifacts) came to dwell solely in the lexicon. Why have literary and cultural studies produced countless treatments of words (the meaningful units of discourse) while lacking a minimally plausible account of grammar (the means of combination)? Later on I’ll say why I think these initial formulations of the question are incorrect or at least imprecise. But first I want to take a step back to answer a more basic objection:

Why shouldn’t the study and theory of literature and culture carry on as it is, untroubled by its lexical exclusivism? Studying words – their history, their use, their social and poetic function – really is a rich and revealing and wholly legitimate pursuit, one that teaches us a great deal about how, in Stuart Hall’s terms, we “impose intelligibility on the world.” If that leaves those who study textual artifacts with a relatively unsophisticated understanding of language, well, sophistication for its own sake is hardly a virtue. We should be comfortable studying keywords (and perhaps a few “small words” or “function words”) and leave the rather arcane aspects of grammar to the linguists who are professionally equipped to deal with them.

It’s true that disciplinary division of inquiry is a rational response to the specialization of knowledge. And oh wow is linguistics specialized. As the linguist Geoff Pullum puts it in his account of the “divorce” of English linguistics from English literature, work in theoretical syntax since the 1960s has produced “over-hyped pseudoscience of a sort that seems almost calculated to be unintelligible to humanities academics, especially when no real scientific gains seem to be emerging from such work” (2010). I admit that my interests have led me to invest substantial time and intellectual labor in getting a handle on X-bar theory and learning to distinguish (for example) raising from control .

But it’s unclear what the payoff would be for understanding and interpreting particular texts, cultures, or histories, since in their very formulation these concepts aim to describe human universals – abstract rules or principles that are subtracted from differences of culture and history. Learning the ins and outs of c-command, or the pros and cons of the DP hypothesis, isn’t going to help me interpret Paradise Lost differently or change the way I understand literary culture, sectarian strife, or political violence in the seventeenth century. And many of these concepts are, so to speak, “theory internal” anyway; they are specific to a particular flavor of grammatical theory; when that theory changes, or another is adopted, they can vanish with a puff of smoke or transform into something quite different. So why bother?

That’s the academic side of things. In popular discourse, by contrast, grammar and syntax usually appear in the guise of nitpicky injunctions issued by judgy pedants and cultural gatekeepers. As long as humanists understand “grammar” to mean debates over the permissibility of split infinitives or dangling participles, they will rightly avert their gaze, devoting their limited attention instead to the word-concepts that articulate our identities, voice our lived experience, and shape our ways of living together. Debating how best to use a loaded word like “culture” can be deeply productive and revealing; debating whether “which” can be used to introduce a restrictive relative clause – not so much.

These objections have force. But there are, I think, compelling reasons why the study of culture shouldn’t stop at the shores of the lexicon. Without an account of a language’s combinatory capacities, we won’t really understand how language articulates and makes the world meaningful. Any account of poiesis (how literary artifacts are made) or hermeneutics (how we interpret texts) limited to words – to their selection and combination – will be inadequate. So long as our theories of the sign travel only the road leading from word to world and back, our accounts of both form and meaning will be hopelessly parochial. I will return to these points another time, but let me focus for now on one deficit in particular. To the extent that we think about meaningful non-linguistic systems by analogy to language – the unconscious is structured like a language; the forces and relations of production are structured like a language; “race is more like a language, than it is like the way in which we are biologically constituted”; culture is like a language and its meanings reside largely in and as language, etc. – the lexicalism of literary and cultural studies puts a check on our analysis of culture, our ability to see and critique how it works.

By way of illustration I want to expand on an analysis of a linguistic form that is treated only briefly in Cyberformalism: Histories of Linguistic Forms in the Digital Archive , which is just out this month with Johns Hopkins University Press. Consider the following utterances, located using Google Books:

“And it’s oh! for the days when Men were Men.”

“…when America was America, when people pulled together and made no bones about it.” (Jack Kerouac)

“when America was America; when there was a West; when a man could start with a capital of one hundred and fifty dollars borrowed money and make good.”

“…in early days, when women were men’s mates, when women were women and not just ladies…”

“dogs today are not what they used to be…. There was a time when dogs were dogs…”

“a day when poetry was poetry, as to children cake is cake, whatever its shortcomings.”

These passages include a range of words that are part of what Judith Butler calls our “basic vocabularies”: men, women, America, dogs, poetry, and, er, cake. But their ideological function – the nostalgic desire, the reactionary construction of history, the location of the ideal in a fictional past – is not performed by the words in this vocabulary or any other, but rather by the abstract linguistic form that they fill, which we can initially notate as N [Past Tense Be] N. If we want to analyze that ideological function to see how it works or to contest its working, then attending to vocabulary, studying word histories with the methods of philology, or deconstructing their binary oppositions, won’t get us very far.

The form N [Past Tense Be] N is part of a family of “tautologous constructions,” a set of related sub-constructions. (This shouldn’t be confused with the classical rhetorical trope, tautologia, which refers to the useless restatement of the same thing in different words.) Other members include phrasal templates and the fixed expressions that instantiate them: N [Future Tense Be] N, as in Boys will be boys; [Art] N be [Art] N, as in A promise is a promise, and most obviously N is N, as in Cake is cake, or War is war, or Kipling’s “East is East and West is West.” If you allow for embedding or repetition (or both), then the family could include Stein’s “A rose is a rose is a rose is a rose,” and Lin-Manuel Miranda’s Sonnet, “Love is love” or its line, “Love is love is love is love is love is love is love is love.” The open categories of these forms – which I will henceforth refer to using the wily spelling of Joyce’s Wake: “tautaulogy – are typically filled by individual nouns, but they accommodate noun phrases (NPs), as suggested by [Art] N Be [Art] N, where [Art] N is a NP, or the now ubiquitous tagline, It is what it is, where [what it is] is a NP.

Love is Love.jpg(The family of tautaulogies is considerably bigger than the sub-constructions listed above, with more distant cousins including I’ll see you if I see you and If it’s broke, it’s broke and qué será, sera and Let Bartlett be Bartlett, and many more. But cataloging must stop somewhere.)

Once you learn forms like these, you can fill them in various ways (potentially infinite ways, if embedded NPs are included) to produce utterances that are new to you and quite possibly (why doubt it?) new in the history of the world. Try it out for yourself!

Tautaulogies are partially unfilled forms that allow you to say new things by filling their blanks in new ways, but they also constrain the ways that their blanks can be acceptably filled. You can announce that Enough is enough, or hark back to the good old days when enough was enough, but saying Much is much or Enough will be enough will likely raise eyebrows. The boys are the boys might not sound wrong, exactly, but I, for one, would be hard pressed to assign it an interpretation, and A boys was a boys will lead people to suspect you’re a) not a competent speaker, b) writing avant-garde poetry, c) a poorly programmed bot, or d) all of the above. (As I said in an earlier post: there’s no end to norms in language.)

Tautaulogies have presented a puzzle to linguists. The problem is that they are apparently vacuous. To the extent that they resemble the logical formula P=P, they seem to express analytic – and therefore necessary – truths. (In the terms of modal semantics, we could say that a sentence like War is War is true in all possible worlds and so doesn’t tell us anything about our world (Snider). And unlike definitions, such as A bachelor is an unmarried man, which are also analytic, tautaulogies don’t even analyze the first term in different terms. So they would seem, on their face, to be completely uninformative.

Yet speakers make felicitous use of tautaulogies with some frequency and even seem to regard them as conveying profound wisdom. And in fact we can give fairly precise and cogent accounts of what tautautologies mean. As Anna Wierzbicka argued in a groundbreaking article, future tense tautaulogies, such as Boys will be boys, express what she calls “a tolerant and indulgent attitude,” and I might instead call an attitude of acquiescence, helplessness, acceptance, or resignation. Tautaulogies of the form [Art] N be [Art] N express obligation in spite of one’s preferences, as when we say The law is the law, A promise is a promise, A deal is a deal, The rules are the rules. This sub-construction is polysemous: other instances, such as A goal is a goal, and War is War, suggest indifferentiation between members of a category: all goals count as goals, regardless of how they are scored, and all wars inevitably share the characteristics of warfare.

Tautaulogies mark differences as well as erase them. Wierzbicka cites the opening words of a Ku Klux Klan leader, “White is white,” and writes that

In such sentences, the speaker stresses the unique quality of something which must be accepted because it cannot be expected to change. But qualities like ‘whiteness’ are also seen as belonging to certain contrastive sets; so, by stressing their uniqueness, the speaker emphasizes the irreducible difference among the members of the set. (98-99)

So tautaulogies can signify the indifferent belonging of members to a group, their contrastive and irreducible difference from members of an opposed group – or both at once. Kipling’s “East is East and West is West, and never the twain shall meet” makes explicit the contrastive set, while “White is white” leaves it implicit. In the semantics Wierzbicka proposes, tautological constructions encode specific attitudes, and the words that fill them specify the categories of things towards which those attitudes are expressed.

Earlier attempts at explaining why it is that tautologies are meaningful despite their apparent vacuity drew on Paul Grice’s notion of implicature, and particularly his maxim of quantity: “Make your contribution no less, and no more, informative than required.” On these accounts, tautaulogies do not say anything informative, but they do imply an informative meaning through the logic of conversation. Because tautaulogies appear to flout the maxim of quantity – uninformative counts as less informative than required – hearers will seek to make them informative by reasoning their way to an unspoken meaning that Grice called a conversational implicature.

Wierzbicka, I think, demolishes the simplest version of the pragmatic account of how tautaulogies mean, which postulates a set of language-invariant maxims (or, mutatis mutandis, a human tendency to maximize relevance) by which speakers reason their way to an implied meaning. This version is unable to account for, on the one hand, the differing semantics of the various tautaulogies within a language (Boys will be boys expresses a different meaning than Boys are boys, while The boys are the boys doesn’t have an obvious interpretation), or, on the other hand, the different semantics and distribution of tautautologies in different languages (Wierzbicka compares English, German, French, Russian, and Polish, and Japanese). In rejecting the pragmatic explanation, Wierzbicka concludes that tautological constructions are language-specific signs – conventionally established pairings of signifier and signified, form and meaning. The meanings paired with tautaulogical forms vary significantly across languages, and speaker-listeners must learn those meanings as part of learning a language. In Russian, for example, tautaulogies indicate that one must do something rather than that one must put up with something. In French one wouldn’t say La guerre est la guerre, though one could say C’est la guerre or La guerre, c’est la guerre. Subsequent studies have shown the wide variation in the conventional meaning of tautaulogies in various languages, including Japanese (Okamoto), Korean (Kwon), Chinese (Wen), and Jordanian Arabic (Farghal).

While Wierzbicka shows that tautological constructions can’t be “FULLY worked out on the basis of any language-independent principles” (99), she does not deny that there are some aspects of their use and meaning that are not conventional and construction-specific. The meaning of a tautaulogy is not wholly arbitrary; it is partly motivated both by its lexical constituents (the tense of be and the characteristics of the repeated NPs) and its parent constructions (more general constructions, such as predication, whose semantic properties tautaulogies inherit). Even once we’ve learned their conventional meaning, comprehending the meaning of their instantiations involves some process of inference, prompted and constrained by convention. Tautaulogies thus present a puzzle of sorts to speaker-listener-writer-readers as well as linguists.

Consider an instance of the sub-construction we started with, “there was a time when dogs were dogs.”  Interpreting this tautaulogy involves introducing difference at the level of the signified even when there is identity at the level of the signifier. We construe dogs as an empirical designation in the subject and as an ideal in the predicate, as meaning, in effect: when those things we call dogs were what dogs should be (Rhodes). (Here I have some doubt about the terms: I might also say that dogs appears first as a cover term, then as an emphatic term.) The nostalgic, reactionary function of the tautaulogy arises from the displacement of dogs = dogs with dogs were dogs, implying that dogs are no longer dogs and, consequently, that now dogs ≠ dogs. In this respect the past-tense tautaulogy shares semantic properties not just with other tautological sub-constructions, but with also with contradictory constructions, such as “Love is not love” from Shakespeare’s Sonnet 116 or Iago’s “I am not what I am,” which mark the semantic/functional difference between the identical terms with a sign of negation, not. But I suppose that the interpretation of many (and perhaps all) tautaulogies involves just this sort of convention-constrained introduction of difference at the level of the signified to supplement the lack of difference in the signifier.

There’s plenty more to say about the linguistic characteristics of tautaulogies – about their semantics, pragmatics, syntax, and so on – but I’ve said enough for now to move on to establishing their relevance to cultural critique. No less than the words of our basic vocabularies, they are linguistic signs that take their place within a larger system of differences. Their meanings are at least in part conventional and specific to different languages. And they are productive, affording speakers and writers the constrained creativity they need to say acceptable things that they have not read or heard before.

Like the words of our basic vocabularies, tautaulogies don’t signify in some neutral sense. They serve a particular ideological function, smuggling in synthetic judgments under the guise of analyticity, giving to culturally specific and conventional judgments the appearance, at the level of the signifier, of necessary and self-evident truth. In the terms of modal semantics, they appear to index all possible worlds while covertly indexing only a subset of those worlds (of which our world may or may not be one). They invite us to perform the convention-conditioned process of inference, even as they make the results of the inference process appear natural and universal. How, without appearing to contradict oneself, can one deny, of anything in particular or of everything in general, that It is what it is?

The semantics of tautaulogies varies by language, but in English, at least, they convey an attitudinal posture of acquiescence or helplessness towards the world as it is, even while measuring the empiricity of the world against pre-given, idealized essences. We can see this ideological function operating in the specific, subject-interpellating domains of gender (Boys will be boys), empire (“East is East and West is West”), nation (When America was America) and race (“White is white”). But their ideological function is not limited to individual or communal subject-formation. In some cases tautaulogies frame acquiescence to existing institutional frameworks, even callous or cruel frameworks, not merely as helplessness or tolerance but as obligation (The law is the law, A deal is a deal). In others they nostalgically locate an ideal in the past and cast the present as a declension from that ideal (When America was America).

To study tautautologies is to study a consequential element of what Gramsci called “common sense” or “good sense,” the popular thought, built from the fossilized fragments of philosophical discourse, that “creates a folklore of the future” – in this case, a future that is bound, irrespective of human agency, to look indifferently like the present (War is war, Boys will be boys). I invoke Gramsci here because he is one of the thinkers who teaches us to see the way that systems of meaning, and especially the meaningful elements of common sense, do not merely function as unequivocal elements of domination, but also serve as resources out of which resistance and dissent are built. This is as true of tautaulogies as it is of words. They are not implements intrinsically tethered to a single class or political interest. Even when they have not been subject to explicit critique, they have been the objects of parody, as when Henry Louis Gates Jr. writes of the “antebellum esthetic position” that wishes to return to a time

when men were men and men were white, when scholar-critics were white men and when women and people of color were voiceless, faceless servants and laborers, pouring tea and filling brandy snifters in the boardrooms of old boys’ clubs.

They can and have been renovated and made critical, taken up and repurposed for oppositional political aims and used as part of attempts to reconfigure the present. Love is love, which appears on protest and yard signs as well as in Lin-Manuel’s poem, shares its semantic and discursive function with other instances of what we can now notate more precisely as NP1 Be NP1, signifying that all members of a category (homosexual and heterosexual love) participate indifferently in that category (love). As with other tautaulogies, Love is love presents its synthetic judgment as natural and analytic, no less necessary that P=P. “War is war,” wrote Gramsci, “and he who embarks on that adventure must feel the full force of the beast that he has awoken.”

Tautaulogies articulate, and partially configure, the contested terrain within movements as well. When the author Chimamanda Ngozi Adichie asserted in an interview that “Trans women are trans women,” she articulated a factional position within the domain of liberal/left feminism. Her utterance shares its semantics and ideological function with other contrastive tautaulogies, such as “East is East, and West is West,” except that it leaves the contrastive pair implict. It unmarks differences within the named group in order to mark differences with another unnamed group, in this case cis women. Of course, it has to be understood not simply as the bearer of its own conventional meaning but as a response to – and indeed a rejection of – another sentence, Trans women are women, often associated with trans activism and trans inclusion, and that appears on posters and yard signs, in newspaper and blog headlines, in hashtags and tweets and social media posts. For purposes of analysis, here are the two utterances:

Trans women are women

Trans women are trans women

It is easy enough to notice that the two statements differ only by the presence of a single word, a single attributive adjective, “trans,” in the predicate NP – a word that also appears in the subject NPs of both statements. But the difference between the two utterances, at the level of form and meaning, can’t be accounted for merely by the presence or absence of a word.  Unpacking the word “trans,” however instructive in other contexts, would contribute little to the analysis of these utterances and their opposition. Trans women are women participates in a non-tautaulogical predicative construction that assigns a species to a genus. Other utterances that participate in the same construction are White women are women and Black women are women and, for that matter, Red cars are cars and Big birds are birds (and so on, quite literally ad infinitum). Though it is not tautaulogical, this predicative construction has, at the level of the signifier, much of the force of a tautaulogy, since the same noun head, “women,” reappears in both subject and predicate, with difference marked only by the modifier in the subject NP. But Adichie’s tautaulogy, “Trans women are trans women,” with identical subject and predicate NPs, competes for the same rhetorical and ideological force. In defining what Judith Butler termed “the subject of feminism,” it unmarks some differences (the differences between one trans women and another) in order to mark others (the differences between trans women and cis women), even as it appears to state only a necessary identity.

To be sure, there is no logical contradiction between the statement that “Trans women are trans women” and Trans women are women.  Yet in a “Clarifying” follow-up Adichie pointedly refuses to say that “Trans women are women,” except to cite it as what is said by another, on the ground that doing so is “disingenuous.” Conversely (and to locate myself) I am unwilling to say that “Trans women are trans women,” except to cite what has been said by Adichie. But this is just to say that the contradiction between the respective statements is a political contradiction rather than a logical one.

I hasten to point out that such statements – slogans, dicta, hashtags, etc. – are not dispositive in issues of ethical, social, and political reasoning, nor should they be. They do not on their own settle ideological struggles or resolve contradictions between real interests (nor have I proposed to resolve them here, as if that were possible). Yet, like the words of our vocabulary, they configure the way that debates unfold. They influence the shapes that factions take and the commitments they assume. Through them factions become legible to one another and, for that matter, to themselves.

Having surveyed the politics of tautaulogies, their role in configuring particular social contradictions, and their peculiar ideological functioning, let me step back to a higher level of abstraction.  In the works of thinkers such as Lacan, Althusser, Barthes, Kristeva, Irigaray, and many, many others, the theory and study of culture adopted its most basic, infrastructural account of meaning and representation from structuralist linguistics. There is no shortage of texts in which that account is rehearsed, but my own reading has most recently included the remarkable lectures on the history of Cultural Studies that Stuart Hall delivered in 1983 and for which the transcripts were recently published. In those lectures, Hall describes how, in Saussure,

the matrix of a language consists of a limited set of elements, which might be as arbitrary as the different phonemic bits of which a language is made up or with which a computer works, and the rules which tell you how to select certain bits and how to combine them with others in order to produce well-formed chains in order to say things intelligibly and correctly. This is an elegantly simple model consisting of elements and rules of selection and combination. (63)

Like many theorists working at the time, Hall silently imports into his exposition of Saussure, Levi-Strauss, and structuralism more generally a number of concepts borrowed from Chomskyan generative grammar. He writes, for example, that “Langue,” a Saussurean concept,

is synchronic deep structure: “deep” because it is not necessarily conscious; “structure” because it expresses the processual and dynamic sequencing of linguistic performance as a static system of rules. (63)

While “synchronic” is Saussurean, the provenance of the term “deep structure” (although not the meaning: see my earlier post) is Chomskyan, as is the concept of “linguistic performance.” Likewise, the notion that “an infinite set of actual speech acts or cultural acts can be performed out of a limited matrix of rules,” is a central doctrine of Chomskyan accounts of language, one that Chomsky himself traces back to the nineteenth century Prussian linguist Wilhelm von Humboldt. The conflation of structuralism with generative grammar aside, Hall’s synoptic history passes from language to myth to “human activities” more generally, including “symbolic systems, social organizations, political systems, or kinship systems” (70), and especially to the “structuralizing” of Marxism: “you have the elements of a mode of production and the different rules of combination through which the forces and relations of production are combined, for example, either in feudalism or capitalism” (100).

Across all of these various domains, structure is composed of two ingredients: “a limited set of elements” and “the rules which tell you how to select certain bits and how to combine them with others.” In the domain of language, we would say: the words of a lexicon and the rules of a grammar. Words and rules, words and rules, words and rules.

Where in the structuralist notion of “structure” composed of words and rules do tautaulogical constructions of the form NP1 Be NP1 fit in? They are simultaneously word-like and rule-like. Like the words in a lexicon, they are conventionally established signs, parings of signifying form and signified meaning. But like the traditional rules of grammar, tautaulogies are means of combination that get filled in creative but constrained ways to produce a large, perhaps indefinitely large, number of distinct and acceptable utterances. This is why I suggested that my starting question – “Why do humanists study only the meaningful units of discourse while lacking a minimally plausible account of the means of combination?” – was imprecisely formulated. Tautaulogies are both meaningful units and means of combination.

So where do they live? One option would be settle them in the margins of the structure, whether as a rule-like supplement to the words of the lexicon or as a sign-like supplement to the rules of grammar. Yet as I argue in Cyberformalism, and as I will suggest in future posts, this settlement is unsatisfactory. Tautaulogies and other linguistic entities like them compel us to reject (in fact deconstruct, though I’ll have to justify that term) the very distinction between words and rules, elements of meaning and means of combination. Following the lead of Construction Grammarians, humanists will need to rebuild, from top to bottom, the “structure” of structuralism – and with it the linguistic foundation of the last half-century of studying, theorizing, and critiquing culture.


The Semantics of “Amazing Racist”


I don’t plan on saying too much here about Randa Jarrar, free speech, or the Fresno State administrators who have threatened to investigate and fire her in spite of tenure. As a tenured professor myself, I have a strong interest in the robustness of tenure’s protections for speech. I am also aghast at the growing tendency of academic administrators to persecute their own faculty at the least sign of controversy, when their role should be to insulate them from a growing tide of coordinated, extramural intimidation. Punishing Jarrar would set a remarkably chilling precedent, one that would discourage other scholars from referring to anyone as a racist. And of course that is the point: there are plenty of people who would be happy either to define the term out of existence or rule it out of bounds, making its very utterance (used accurately or not) grounds for punishment or dismissal. But there are racists; they are real; they have played and continue to play a significant role in public life; and there are both epistemic and ethical reasons for scholars to see them and call them what they are.


Instead of discussing those bigger topics, most of which have been covered from a multitude of different angles in various venues, I want to attend for a moment to the rhetoric, syntax, and semantics of the phrase “amazing racist.”


Here is the phrase in the context of Jarrar’s original tweet:


“Barbara Bush was a generous and smart and amazing racist who, along with her husband, raised a war criminal.”


I find it interesting, for a start, that the phrase “amazing racist” is the segment of the tweet that has gotten bandied about in numerous headlines and quotations, since “amazing” is just one of three coordinated attributive adjectives – “generous,” “smart,” “amazing” – that together modify the noun “racist.” More often than not, adjective sequences like this one are spoken or written without any conjunctions – “a generous, smart, amazing racist” – or with at most one conjunction separating the penultimate from the ultimate adjective. Jarrar’s use of an additional “and” is an example of what classical rhetoricians called polysyndeton, which sounds technical in English, but would have just meant “bound together with many” to the Greek speakers who first used it.


The tweet performs a rhetorical two-step that is part of the dance academics perform daily: praise preceding criticism. On the papers of students and colleagues, an English professor like Jarrar has doubtless written countless comments with just this sequence:


“This essay’s treatment of its topic is imaginative and shows off your abilities as a prose stylist, though it lacks a central thesis statement, quotation of any textual evidence, or paragraph structure.”


The honeyed praise on the rim of the cup makes more palatable the often-bitter medicine of criticism that follows. It can also help to establish the critic as a judicious and balanced and even sympathetic reader. Of course, in her tweet Jarrar isn’t offering judicious and constructive criticism (and Barbara Bush is no longer accepting any). The opening praise has a different purpose, as the rhetorical equivalent of the “deadly hug.” The sudden drop from the heights of “generous and smart and amazing” to the depths of “racist” is a time-honored writerly trick, a way of making the sentence dramatic.When Robert Frost wrote that “Everything written is as good as it is dramatic,” he articulate a principle especially applicable to tenure-endangering tweets.


The two-step of praise and criticism, praise and blame, is usually expressed in distinct, or at least distinguishable propositions, as when, in his conversation with William Drummond, Ben Jonson said that “Samuel Daniel was a good and honest man, had no children; bot no poet.” But Jarrar’s tweet is different: its praise and blame come conjoined in a single noun phrase – “a generous and smart and amazing racist” – with the coordinated adjectives giving the hug, the noun thrusting the dagger.


The conjoining of praise and blame in a single noun phrase brings “a generous and smart and amazing racist” close to the trope of oxymoron, a kind of compressed paradox or union of opposites, of which Milton’s “darkness visible” is probably the most notable in English literature. (From the Greek oxy “sharp” and moron “dull,” oxymoron is a self-exemplifying term – an example of the phenomenon that it refers to). Of the three adjectives, “generous” is arguably the one that most closely approaches an oxymoronic pair with “racist,” though its distance (in the time of reading and the space of the screen) from the noun diminishes the effect of the opposition. But there is, I think, more going on in the phrase than the figurae of classical rhetoric will help us to pick apart.


Readers of Jarrar’s tweet will at first construe the coordinated adjectives of the noun phrase as approbatives and only on encountering the noun have to recast them as less than or other than praise. Of the three, “amazing” undergoes the most dramatic reversal, appearing at first to mean “astounding” or “wonderful,” as it would in the phrase an amazing experience. But it also functions as an intensifier, rather like real, downright, total, or terrific. (If Jarrar had called Bush amazingly racist, with the adverbial form, then the intensifying function would be the first and only one.)


In addition to asserting that she was a racist, was Jarrar also asserting that Bush was generous, smart, and amazing? It’s hard to tell. Semanticists, following the work of Barbara Partee, have distinguished between subsective and intersective adjectives. If you say that Chad is a tremendous douchebag, I would be wrong to think that you’ve asserted that Chad is tremendous. That’s because “tremendous” here is subsective: it picks out just that subset of douchebags who are tremendous as douchebags. By contrast, if I say of my dog, Derby is a loyal and gentle pup, this entails that she is loyal and gentle. If you learn that she’s also a mutt, you would be right to conclude that she is a loyal and gentle mutt (and she is!). When Jonson said that “Samuel Daniel was a good and honest man” he was using intersective adjectives. But If she had tweeted that “Barbara Bush was a generous and smart and amazing woman,” the adjectives would clearly be intersective.  With the noun head “racist,” the tweet is ambiguous between subsective and intersective readings. Is a “generous racist” a racist who is generous? Or is an “amazing” racist amazing only as a racist?  I wonder if the difference between subsective and intersective adjectives comes up in interpreting legal documents.

“Amazing racist” is an evocative phrase not simply because of its ambiguous, whipsaw meaning. It carries within it (and could very well be prompted by) echoes of the hymn “Amazing grace” or of the TV show “The Amazing Race.” (That echo is present as well in the title of the 2015 novel, “The Amazing Racist,” by Chhimi Tenduf-La, and the stage name of the (sucky) comedian Ari Schaffer.) Searching the NOW Corpus at Brigham Young University, which contains “5.9 billion words of data from web-based newspapers and magazines from 2010 to the present time” with the string _j* racist, we can get a sense of what adjectives most frequently modify the noun “racist.”


“Amazing” makes the list at 61, partly because a few articles about Jarrar’s tweet and its consequences have been pulled into the corpus in the last few days.  “Generous” and “smart,” unsurprisingly, don’t appear.


Disciplinary Transformations

Since finishing work on my second book, Cyberformalism, which uses digital tools to expand the domain of philology beyond words to include abstract linguistic forms (partially or fully lexically unspecified constructions like X is the new Y), I’ve been doing research to answer a few questions about the disciplinary formation of literary and cultural studies and the humanities more generally. How did we come to work with an understanding of language narrowed to the lexicon? How did we end up immuring ourselves (with rare exceptions) in the study of what Judith Butler has called our “inherited vocabularies”?  How is it that our books and articles traffic in keywords while lacking anything like an explicit or theorized account of the combinatory and creative aspects of language use?

Though the primary aim of Cyberformalism is to show how we might use digital methods to begin to fill this gap, the book also offers the outlines of an answer to this question, one that involves the combination of print research tools (dictionaries, indexes, concordances, catalogs) and longstanding assumptions about the building blocks of language – and in the last century about the nature of the Saussurean sign. While I think what it proposes is basically right, there’s much more to say about the disciplinary history that lulled scholarship on literature, culture, and intellectual history into its current lexicodogmatic slumbers.

In seeking to understand the grammar shaped gap in our understanding of language, I’ve been reading work from the 60s and 70s (often but not always published under the banner of “stylistics”) that saw in Chomskyan grammar an opportunity for understanding how authors made, and readers comprehended and experienced, sentences they’d never encountered before. Most recently I spent time with Ann Banfield’s 1973 dissertation, Stylistic Transformations: A Study Based on the Syntax of Paradise Lost, written in partial fulfillment of the degree requirements for a doctorate in English at the University of Wisconsin, Madison two years before joining the English Department at UC Berkeley, where she spent her teaching and research career and is now professor emerita.

The dissertation is, for many reasons, a remarkable document, both in itself and what it says about its moment in the history of the discipline. None of it, so far as I can tell, made its way into print. Having done a fair bit of research on Milton’s style, my sense is that it is cited barely (if at all) in subsequent work, though it is easily the most systematic treatment of Milton’s grammar and its relation to style. (I did not discover it in time to cite it in the relevant chapter of Cyberformalism.) Its limited influence on subsequent scholarship may also be a result of Banfield’s unusual career. She went on to publish influential arguments on narrative theory, free indirect discourse, modernism, Beckett, the Lacanian theory of the subject, and much else… but not Paradise Lost. In her first book, published seven years later, she described herself not as a critic or historian or scholar of literature but as “a linguist working within the Chomskyan paradigm” (Unspeakable Sentences, ix), though literary works continued to be the objects of her analysis over the subsequent decades of an accomplished career.

In its rigor, clarity, and expertise, Stylistic Transformations is an impressive piece of scholarship and an especially masterful feat for a doctoral student. It is also an artifact from a lost age. The expertise it exhibits could hardly be more different from what is now expected of current literature PhDs. It is uninterested in the history, the context, or even – in all but peripheral cases – the meaning of Milton’s poem. Though it sporadically cites a few prominent early readers and editors of Paradise Lost – Johnson, Addison, Bentley – it otherwise has no archival basis. Its central thesis is not about Milton’s epic at all, which serves instead as an exemplary text – a corpus of sentences – on which to develop a sound “theory of stylistic deviance” (vii), one that does not depend on the historical and contextual vagaries of a statistically conceived norm or require developing an “independent stylistic grammar” for each poet or poem (xii).

Adopting the post-Aspects (1965) transformational grammar of Noam Chomsky, Banfield identifies a stable, invariant stylistic norm, generated by the “phrase structure rules” of the syntactic “base” (xxii), which produce simple, active, declarative sentences followed by a set order of complements, and from which the sentences in the poem can be analyzed as deviations. These deviations, she suggests, are not merely stylistic, but in fact are, to varying degrees, “actual departures from strictly grammatical English.” She proposes that native speakers of English, “when presented with a passage from Paradise Lost, will not only recognize Milton’s departures from normal English word order, but will also be able, with surprising consensus, to reorder the passage to correspond to normal English word order” (5), though she demurs from equating this norm, which she also calls “the normal order of major constituents in a sentence,” with Chomsky’s controversial notion of “deep structure” (3). The first and longest illustration of a sentence “reordered to conform to normal English word order” is especially instructive. Here’s Milton’s verse:

Now had the Almighty Father from above,
From the pure Empyrean where he sits
High Thron’d above all highth, bent down his eye,
His own works and their works at once to view:
About him all the Sanctities of Heaven
Stood thick as Starrs, and from his sight receiv’d
Beatitude past utterance; on his right
The radiant image of his Glory sat,
His onely Son; On Earth he first beheld
Our two first Parents, yet the onely two
Of mankind, in the happie Garden plac’t, (Paradise Lost 3.56-66)

And here is Banfield’s reordering:

The Almighty Father had now bent down his eye from above, from the pure Empyrean where he sits high Thron’d above all highth to view His own works and their works at once: all the Sanctities of Heaven stood thick as Stars about him, and received Beatitude past utterance from his sight; the radiant image of his Glory, His only Son, sat on his right; he first beheld our two first Parents on Earth, yet the only two of mankind plac’t in the happy Garden. (5-6)

The reordering procedure is not implausible on its face, though one could propose alternative reorderings that would be no less plausible (i.e., “bent down his eye to view his own works… from above, from the pure Empyrean…”).  Banfield makes no mention of the pursuit of human biological universals that is central to the Chomskyan paradigm. Yet in establishing the reordering procedure she seeks to factor out historical difference to every possible extent. “Milton’s English, in its essential features, is Modern English,” such that the “native-speaker intuitions of the modern reader of Paradise Lost” are the “most helpful guide to the grammatical status” and stylistic deviations of the poem (xxi). Its sentences are not “examples of acceptable constructions that have fallen out of use since Milton’s day”; they are, to varying degrees, as ungrammatical for Milton’s original audience as they are for present day readers.  Our norms are his norms.

Having established the reordering procedure that reveals the invariant, intuitively available norm, Banfield goes on to consider how it is that the “deviant constructions” (224) of Milton’s poem are derived from it. She adopts John R. Ross’s conception of “scrambling,” which produces the “free word order” of inflected languages such as Latin, from the underlying phrase structure output of the base rules, but she qualifies that in English – and specifically the English of Milton’s grand style – the “scrambling” rules require some modifications to prevent them from generating fully ungrammatical and intolerably ambiguous constructions.

The bulk of the dissertation is a study of precisely what those modifications might be and what transformations are permitted. It is carried out in technically virtuosic analyses that categorize a quite large number of sentences from Milton’s epic, some of them complete with phrase structure markings, tree charts, and transformation diagrams.

Banfield Syntax 2.jpeg

What sorts of transformations – movements, deletions – from normal constituent order are permitted? Which of the usual grammatical constraints are relaxed or flouted? Here Banfield shows a remarkable grasp of the state of the field of linguistics and literary stylistics circa 1970, drawing on a wide range of hypothesis proposed by many early transformationalists. To the two components of syntax, base and transformation, established by Chomsky, Banfield proposes adding a third, distinct “stylistic component.” The stylistic transformations of this component, which produce the “departures from strict grammaticality” characteristic of the grand style, are “late rules” that apply only to the “output of grammatical rules” (225). Milton’s sentences are thus the product of three successive operations: the first generating the simple, active, declarative constructions; the second transforming these constructions to produce questions, passives, subject-auxiliary inversions, and the like; and the third performing a subsequent round of transformations to produce the semi-grammatical peculiarities of what Dr. Johnson termed Milton’s “Babylonish Dialect.”

Page after page, I found myself marveling at the virtuosity of Banfield’s analysis, even while finding little reason to accept the theoretical paradigm in which it operates. Trained up in literary studies some four decades later, I am disposed to be skeptical of posited “norms” – whether biologically inherited, as in Chomsky, or simply evident from “native-speakers intuitions” (vi), as in Banfield.  The term “skepticism,” as I use it here, is not eliminative.  It does not suggest the nonexistence of the norm: surely there are norms, no end to norms, in language.  Instead it poses a range of historical and constructivist questions: How did this norm (the simple, declarative sentence, followed by a set order of complements) become the norm?  What made it available to “intuitition”? How does it vary across historical time, linguistic cultures, and even specific circumstances?  What role did ostensibly “deviate” transformations – putatively “late” rules – play in establishing it?  For most of the last 40 years, this sort of disciplinary skepticism has been primarily (and for good reason) trained on norms regarding raced, gendered, classed, and differently abled bodies.  But it has methodological purchase even in the lower-stakes domain of grammatical theory.

(It is worth noting that Chomsky himself, even as the nativist aims of his grammatical theory have remained largely unchanged, has moved on from the transformational paradigm to the current Minimalist Program, which, roughly, replaces the two distinct syntactic components, base and transformation, with two flavors of a single operation, merge, that forms labeled binary sets from individual constituents.  Because there is no syntactic base and no semantically determinative “deep structure,” the treatment of norms are quite different, in ways that I won’t – and am not really qualified – to explore here.)

Despite my skepticism of its method, Banfield’s dissertation offers insight into the strange and sublime experience of reading Milton’s sentences with a degree of rigor, formal explicitness, and comprehensiveness that is not to be found either in earlier treatments by Christopher Ricks or later ones by Tom Corns or John Hale.  (These treatments have other virtues that make them valuable.) She establishes a systematic framework within which to answer questions – how did Milton make this poem, sentence by remarkable sentence? what must he have known (probably unconsciously) to do so? – that subsequent literary scholarship, following its abandonment of the Chomskyan paradigm, has lacked the intellectual resources to ask in a coherent way.

Banfield’s dissertation, and her subsequent book, Unspeakable Sentences, represent high water marks in the synthesis of grammar and literary studies.  In future I’ll be exploring – and probably writing about here – why that synthesis fell apart so fully by the end of the 1970s. The Berkeley English Department may end up serving, if not exactly as the institutional home of this disciplinary transformation, then as a sort of main stage on which it unfolded. In the 60s and 70s at Berkeley, Donald Freeman and Seymour Chatman edited volumes including essays by literary scholars alongside noted generative linguists, such as Morris Halle and Paul Kiparsky. It was there that Stanley Fish published his first attack on stylistics (which was also an attack on Chomskyan linguistics) in 1973, the same year that Banfield filed her dissertation.  Stephen Greenblatt was already teaching at Berkeley when Banfield arrived and would published “Learning to Curse,” an early exemplar of what would come to be known as the New Historicism, a year later.  Not all of the relevant history took place at Berkeley, of course.  The dramatis personae of my further research will probably include Annabel Patterson and Richard Ohmann.  And it will need to attend as well to the shifting configurations of theoretical linguistics, which since the 70s has been more prone to seek alliances with neuroscience, evolutionary biology, psychology and even physics than literature or history.