Form and formalism in linguistics

Dennett, Daniel. 1995. Darwin’s Dangerous Idea: Evolution and the Meanings of Life. New York: Simon & Schuster; London: Allen Lane. Müller, Friedrich Max. 1861. Lectures on the Science of Language. [First series.]. London: Longman, Green, Longman & Roberts. Müller, Friedrich Max. 1873. “Lectures on Mr Darwin’s Philosophy of Language.” Fraser’s Magazine, 7.525-541, 7.659-678, 8.1-24. Piattelli, Michela. 2016. “‘Language Is Our Rubicon’: Friedrich Max Müller’s Quarrel with Hensleigh Wedgwood.” Publications of the English Goethe Society 85 (2–3): 98–109. Volume republ. as Friedrich Max Müller and the Role of Philology in Victorian Thought, ed. by John R. Davis & Angus Nicholls, Abingdon & New York: Routledge, 2018. doi:10.1080/09593683.2016.1224511. Radick, Gregory. 2000. “Language, Brain Function, and Human Origins in the Victorian Debates over Evolution.” Studies in History and Philosophy of Biological and Biomedical Sciences 31 (1): 55–75. doi:10.1016/S1369-8486(99)00045-X. Wedgwood, Hensleigh. 1848. On the Development of the Understanding. Taylor & Walton: London. Wedgwood, Hensleigh. 1859-65. A Dictionary of English Etymology. Vol. 3. 2nd ed. 1872, 3rd ed. 1878 ed. London: Trübner. Wedgwood, Hensleigh. 1866. On the Origin of Language. London: Trübner.

First things first: don't miss the digital price. It's free! The hardcover is handsome, stitch-bound, reads beautifully, is extraordinarily well priced, and even has the sort of built-in string bookmark usually reserved for bibles and classical poetry. I'm happy to have it on my shelf, at close reach. But the digital copy is an offer you can't refuse. 1 Not everything free is worth the price, however, and the real cost of a book, whatever its assigned monetary value, is paid in time. Let me assure you: Form and Formalism in Linguistics abundantly repays hours and hours of investment.
In a brief review of a rich and diverse collection like this one has to play favourites or else confine oneself to a few sentences for each chapter, doing little more than expanding the titles or shrinking the abstracts. I play favourites. But all of the chapters are rewarding. The scholarship is routinely first-rate, the prose ranges from good to graceful, and the historical narratives woven among the various analyses are consistently engaging. I recommend them all. So, before I discuss a few favoured chapters, let's itemise the contents. In addition to a brief and exemplary preface by the editor that epitomises and contextualises the chapters, the following nine studies are included: Visual formalisms in comparative-historical linguistics, Judith Kaplan Alternating sounds and the formal franchise in phonology, James McElvenny On Sapir's notion of form/pattern and its aesthetic background, Jean-Michel Fortis Linguistics as a "special science:" A comparison of Sapir and Fodor, Els Effers The impact of Russian formalism on linguistic structuralism, Bart Karstens The resistant embrace of formalism in the work of Émile Benveniste and Aurélien Sauvageot, John E. Joseph Linguistics as a science of structure, Ryan M. Nefdt Formalism, grammatical rules, and normativity, Geoffrey K. Pullum Linguistic form: A political epistemology, Nick Riemer The coverage is limited. There's barely a hint about iconicity, for instance, though there's only so much that can get done in 271 pages. 2 But one important notion about form that is routinely ignored in linguistics gets some welcome attention here, the allowance that linguistic form might be aesthetic in some fundamental way; that is, not just when it shows up in special registers like epic poetry and hip hop. That idea is implicit in Bart Karsten's chapter on Russian Formalists' influence on structuralism, but it gets a much fuller airing in Jean-Michel Fortis's 'On Sapir's notion of form/pattern and its aesthetic background'.
Sapir was motivated by a sense of 'rightness' in linguistic patterns, what he often called 'form-feeling'. Fortis traces this sense directly to Sapir's musical interests, to aesthetic theory, and to the Gestaltists. Empirically investigating and formalising such general notions as similarity, contrast, and contiguity, which have attended psychology from at least Aristotle, the Gestaltists showed that we intuitively feel some patterns to be more natural than others. We categorise percepts rather automatically as 'belonging' with each other, or not, based on features like size, shape, colour, and proximity, assembling those percepts into larger, integrative patterns. As Sapir puts it in a beautiful analogy about phonological systems, the assembled pattern then defines the appropriateness and function of any given element: A sound that is not unconsciously felt as "placed" with reference to other sounds is no more a true element of speech than a lifting of the foot is a dance step unless it can be "placed" with reference to other movements that help to define the "dance". (Sapir [1924(Sapir [ ] 1951 (64) Sapir's observation, too, that 'all languages evince a curious instinct' to over-ride functional demands on form by the 'sheer play of [their] means of expression' (1921: 60) (72) is highly consonant with the growing linguistic interest in the playful and 'poetic' aspects of language (Carter 2004;Dancygier & Sweetser 2014;Benczes 2019).
A book on linguistic formalism would be unthinkable without the looming presence of Noam Chomsky. He has walk-on parts throughout, a supporting role in a couple of chapters, and a starring role in one. The most interesting of his supporting roles comes from the pen of that always interesting linguist, Geoffrey K. Pullum.
Pullum's chapter carefully probes Chomsky's scholarly debt to Emil Post, the brilliant mathematical logician who developed production systems. In 'Chomsky's hands', Pullum reminds us, 'production systems, under the new name "generative grammars", became the overwhelmingly dominant type of framework for the study of syntax' (202). But Chomsky only referred to Post's work in brief and scattered remarks, never citing the paper in which Post published his primary results. On the surface, this looks to be either remarkably negligent or sneakily dishonest, two descriptions that have frequently been applied to Chomsky's practices. Pullum cares about academic integrity, and is not exactly the president of Cheerleaders for Chomsky, so both hypotheses get a full hearing. I'll be coy, as Pullum was in a related talk for the 2019 Henry Sweet Society meeting, and encourage you to read the chapter to get his verdict. Doing so, you'll also encounter fascinating (even, for some, provocative) suggestions about the relations among rules, normativity, prescriptivism, and grammaticality.
Ryan M. Nefdt's chapter, the one in which Chomsky has a starring role, pairs well with Pullum's. 3 Nefdt traces generative grammar post-Post, through its half century of radical theory change, 'each new foundation distanc[ing] itself from the methodology of its predecessor, postulat[ing] different objects and advocat[ing] different ends' (178), a situation made all the more puzzling, 'since practitioners of each epoch of the theory can still be found working within the remit of their chosen formalism' (185). Nefdt reads this history against what is widely known in philosophy of science as the pessimistic meta-induction (PMI) -namely that, since all previous theories of x have proven false or inadequate, there is no justification to put any stock in the current theory of x (or, more generally, science can't be describing or explaining really real reality since its theories always turn out wrong in the end). The chapter traces out the originary generative-grammar tandem of transformational and phrase structure rules through their mutating formalisms as Chomksy shifted various allegiances among expressivity, learnability, complexity, simplicity, evolutionary plausibility, and so much more, until their eventual, current, merger into Merge. (Understandably, Nefdt does not attempt to trace these rule types and their formalisms in their various non-Chomskyan generative mutations.) The argument is curious, offered as a kind of vindication of Chomsky's program in light of the PMI. The pivot of Nefdt's case is that changes of commitments and formalisms have not affected the structural core of the Chomskyan framework because certain familiar features of his various mechanisms (weak generative capacity, infinite output from finite input, and hierarchical representation) have remained stable throughout. This '[ontic] structural realism', according to Nefdt, gives us reason to place our faith in generative grammar, by tethering the program to the structures it has consistently posited. But as the title suggests, which doesn't mention Chomsky, transformations, or phrase structure rules, the Chomskyan program is held to be representative. It is linguistics as a whole, as 'a science of structure', that structural realism saves from the PMI, and, in a brief coda, Nefdt extends this idea of continuous core features to earlier frameworks, back through the Bloomfieldians and Firthians to at least Saussure.
This extension would seem to include Émile Benveniste and Aurélien Sauvageot, whose '"resistant embrace[s]" [of] structuralist formalism' (141) are the subject of John E. Joseph's absorbing chapter. Joseph uses his oxymoron not just to chart the ways Benveniste and Sauvageot held their noses while hugging structuralism, but also to chart how two enduring tensions shaped linguistics in the period. It was problematic for linguists at the time to forebear embracing structuralism in public. It was the terminological fashion of the day. If you reject the mainstream, you can expect the mainstream 3 There is, however, a rather curious editorial slippage in the Nefdt chapter with respect to Pullum,in that Nefdt (191) cites a forthcoming work by Pullum (now out, 2019) to establish a point that Pullum makes in the very next chapter (201).
to reject you as well, and your ideas, and your students, who might need access to publications, jobs, students of their own. So, you dress in the fashion of the day. But wearing the uniform does not always mean drinking the Kool-Aid. The resistance in the embraces of Benveniste and Sauvageot had multiple and overlapping motivations, chiefly the role of speakers in the formation and evolution of language, which Kool-Aid structuralists ignored, and insufficient attention to linguistic details in structuralist methods. But they both also had reasons beyond the pragmatics of fashion to embrace (some) structuralist practices and (some) structuralist commitments. Benveniste, in particular, was drawn by the appeal of orderliness that governed the movement, manifesting one Joseph's two tensions, between the universalism of prevailing order and particularities of speakers and their cultures. The other tension is between holism and atomism -evident, for instance, in Ernst Cassirer's observation that language 'does not consist of detached, isolated, segregated facts', but 'forms a coherent whole in which all parts are interdependent upon each other ' (1945: 110) (149). Cassirer? I thought this chapter was about Benveniste and Sauvageot? The title is somewhat misleading, actually, as Joseph's cast of characters is quite large, an ensemble that includes, among others, Roman Jakobson, Claude Lévi-Strauss, André Martinet, Antoine Meillet, and Henri Meschonnic, with themes and storylines that implicate political affiliations, national allegiances, friendships, and personal antipathies.
What Joseph explicitly seeks to avoid by unparadigming French structuralism in this way is a history that fits Voltaire's description as 'a pack of tricks we play on the dead'. He frames the period not by its practices and commitments but by its practitioners and adherents. He claims at the end of his essay, in a clever play on a passage from Benveniste, to have been searching for the 'Order which governs the movement of the stars of the modern science of language' (170); in my reading, rather, he has found Order not governing our stars, but arising from them, from their collective-but-notalways-communal labours. Indeed, to varying degrees, that kind of attention to nuance and telling details characterises all of the chapters in the book.