Wednesday, December 30, 2020

Books read 2020

The pandemic caused me to spend large parts of 2020 sitting alone in my flat, so it's not surprising, nor is it a particular achievement, that I read lots of books this year (61 in total). As usual, here are my mini-reviews of all of them.

Donna Beth Ellard, Anglo-Saxon(ist) Pasts, postSaxon Futures

Eclectic and erudite, this book takes us from Ellard’s own upbringing in Mississippi to King Alfred’s disease-ridden body via malt liquor consumption and the nineteenth-century ofermod of people like Thorpe and Kemble. I find it difficult to read and be persuaded by every aspect of this fiercely personal book, but that’s likely my failing rather than the author’s. Worth reading for anyone invested in “Anglo-Saxon Studies” as a discipline.

John Guy, Mary, Queen of Scots

I wish more academics could write like this. This book is a good read at the same time as being unquestionably written by someone at the top of his game academically, and I learned a tremendous amount. Guy is sympathetic to his biographical charge, never to the point of sentimentalizing, but sometimes in a way that strains credulity slightly. But that may well be a necessary reaction to the rather dismissive way Mary tended to be treated in the previous literature.

Rosina Lippi-Green, English with an accent

This is a fantastic book, which should almost be required reading for linguistics undergrads, if it wasn’t so specific to the US context. The issues, of course, are the same everywhere at the right level of abstraction. I’m even tempted to teach a course on this basis at some point. (Me from eight years ago would be shocked.)

Robert McColl Millar, System collapse, system rebirth

Despite the interesting topic (demonstratives and definite articles OE-ME) and great data collected, I didn’t get on well with this book. The writing style is jarring and (dare I say it) verbose, and there’s a lot of subjectivity to his evaluation of previous scholars’ claims. The constant and impersonal talk of linguistic “systems”, as if these were something that existed independently of speakers with their own desires and motivations, is also frustrating to me, and I was hoping that the discussion of the Norse-English contact situation would be longer and more nuanced.

Brigid Kemmerer, A curse so dark and lonely

This fantasy romance, a novel take on the Beauty-and-the-Beast setup, is not my usual fare, and I can’t say I loved it. The writing style is nice and flowing, but the characterization - especially of the protagonists, and especially Rhen - was somehow lacking. 

Douglas Biber & Bethany Gray, Grammatical complexity in academic English: linguistic change in writing

I came to this book to read about genre and register, but I guess I’ll need to read different works by Biber to find out about that. What I did discover is that stereotypes about academic writing are mostly wrong: in particular, embedded clauses are less common in academic writing than in conversation, whereas nominal modification (especially premodification of a noun by another noun) is much more common there. The empirical findings are fascinating, but the theoretical underpinnings come across as weak. The authors attack the idea that linguistic change always originates in speech, but I don’t think they’ve seriously got inside the mindset of people who make this claim, so their attack largely misses the mark. More disturbingly, the science-humanities dichotomy adopted here is seriously naive, especially as regards its historical origins.

N. K. Jemisin, The Stone Sky

A great conclusion to what must be some of the best fantasy writing of the century so far. Some of my quibbles with the previous book carry over to this one: the supporting cast in Essun’s part of the story are drab, and in fact that whole storyline is pretty tedious (I get why it’s there, but it doesn’t have the desired effect on me). The worldbuilding and in particular the ending are spectacular, though.

Peter Heather, The Goths

Probably still the go-to read on the Goths, focusing on when we know most about them, from the late fourth to early sixth centuries. Heather is a counter-revisionist in matters migration and ethnicity, and where there is more critical discussion it comes down on one side or the other, though I can’t help thinking that he’s too eager to invoke ethnicity per se rather than other kinds of group identity.

Irene Heim & Angelika Kratzer, Semantics in Generative Grammar

Formal semantics and I have a love-hate relationship, if it’s possible to have such a relationship with something you don’t fully understand. I don’t really see how or why this sort of theory is about meaning at all, rather than being another kind of syntactic theory which we then have to rather clunkily map onto a different syntactic theory. But either way this theory makes some odd predictions. Quantified nominals, for instance, come out as radically different from other kinds of nominals - something which, as far as I’m aware, there’s zero evidence for in natural language. And the book spends a long time trying to justify these predictions, and posit weird workarounds to get them to map onto the normal syntax assumed for them. Perhaps more importantly for my perspective, I read this book in the hope of finally understanding lambdas, but here too I was disappointed. It would help an awful lot if the many exercises in this book actually came with answers provided.

Jürgen Udolph, Ostern - Geschichte eines Wortes

There are two traditional stories for the etymology of “Easter” (German Ostern). One, going back to Bede and Jakob Grimm, claims the existence of a goddess Eostre/Ostara. The other links it with the dawn. Udolph argues that both are wrong and that the word goes back to a word ausa related to baptism. A cute little book - no idea how seriously it’d be taken by true etymologists, especially since it’s light on real linguistic detail (sound changes, etc.).

Richard Watts & Peter Trudgill (eds.), Alternative histories of English

A much-needed book with some glaring limitations. Fantastic on different varieties of English and on ideologies. But virtually all the chapters focus mainly or exclusively on the post-1500 period - is there no alternative before that? And James Milroy’s chapter is awful. But I’ll try to make good use of the rest in the textbook I’m writing and in my teaching.

Hilary Mantel, Wolf Hall

Follows Thomas Cromwell, Henry VIII’s fixer. Since this book has won virtually every award in existence there’s little I can add. The prose style is beautiful at the micro-level. At the macro-level it’s hard to figure out where things are going, which conversations are important, etc. And Mantel’s use of pronouns is peculiar.

D&D Dungeon Master’s Guide

Of the three D&D core books (also including Player’s Handbook and Monster Manual) this is the least useful. That’s not to say it wasn’t worth reading. Even the least imaginative DM will be able to be creative using this book - you can randomly generate virtually everything: items, NPCs, dungeons, even villains’ motivations. And I got a few ideas for the campaign I’m running.

Enoch Aboh, The emergence of hybrid grammars

Despite the title, this is basically a contribution to the ongoing “creole debate”, from a generative perspective. I learned the most from chapter 2, which deals with the sociohistorical circumstances of the Slave Coast. Chapter 3 critiques existing theories of creole origins, and chapter 4 proposes an alternative. Chapters 5, 6 and 7 are then in-depth syntactic analyses of Saramaccan, Haitian Creole, and their ingredients English, French, and Gbe in particular. I’m perhaps not as sympathetic to the book’s proposals as I should be. “Recombination” is very unconstrained; Aboh doesn’t like reactions to earlier versions of this work who’ve described it as “mixing”, but it’s not clear whether this difference is contentful. Aboh appeals to the (in my view awful) Interface Hypothesis to motivate why some things are vulnerable and others not, but this is probably the weakest part of the book. More generally, I’m just not convinced that Aboh’s method of in-depth analyses of particular syntactic constructions is suited to answering questions about creole exceptionality, regardless of its more general merit. To bring data to bear on a general typological claim it’s just not sufficient to look at data from two creoles, unless the claim can be straightforwardly falsified by counterexample. That might be true of (for instance) the claim about contextually sensitive morphology, but it’s not true of McWhorter’s specific claims. Aboh has a barbed rejoinder to McWhorter’s (surfacey) method in chapter 4. And it’s true that lots of data doesn’t help your case if the data is crap. What’s needed is work that is both syntactically rigorous (like Aboh’s) and broad-based (like the typological work) - and until we have that, the two sides of the “debate” will probably inevitably continue talking past one another.

Cixin Liu, Death’s End

The conclusion of a quite incredible trilogy - can’t be faulted for its scope or vision. The mechanisms introduced in this book remind me of Vernor Vinge’s Fire Upon the Deep, though it’s all discussed more explicitly. I didn’t quite like this one as much as the second volume, the Dark Forest, because it was more disjointed; there’s just so much going on in this book. I would have preferred it if Part VI hadn’t been included at all. But it’s still the conclusion of a phenomenal achievement, and I can now file this in the rare category of trilogies whose second member is the best.

Goethe, Wilhelm Meisters Wanderjahre, Erster Teil

It’s always unpleasant to rediscover that you’re just too stupid to understand some books, and so it was for this one. Wilhelm wanders around, meets some people, and there are plenty of little parable-style stories embedded in the larger narrative, mostly about love and mistaken identity. If there’s a larger message I’m missing it entirely. I guess I’ll read part 2 because I’m a completionist, but I’m not hugely looking forward to it.

Martina Wiltschko, The universal structure of categories

If this book had been written by a man and drawn primarily on Romance or Germanic data, it’d be hailed as a magnus opus and cited much more. Its aim is to propose a new theory of how categorization in language works, a kind of “middle way” between cartography with its innumerable innate functional projections and anything-goes typological nihilism. On the whole I suspect it will convince neither camp. The syntax assumed is bafflingly abstract, with loads of null situation pronominals - and few actual syntactic analyses of sentences. Moreover, the notion of “substance” or “substantive content” is not at all clear (at least to me), and the functions proposed for the spine (“anchoring”, “point of view” etc.) seem too vague to really make clear predictions. Though it could be that I’ve misunderstood something dramatically - it wouldn’t be the first time. In any case, an ambitious and thought-provoking read.

Robert Flierman, Pagan, pirate, subject, saint: defining and redefining Saxons, 150-900 A.D.

This PhD thesis deals with how the term “Saxons” is used by outsiders and by Saxons themselves, starting with Roman authors. It’s interesting and well-written, and I learned a lot. Chapter 4 is a bit of an outlier, dealing mostly with the Capitulatio as a source and its dating, rather than its representation of Saxons per se.

George R. R. Martin, Fire and Blood

This book is to the Song of Ice and Fire as the Silmarillion is to Lord of the Rings. Dealing as it does with the early Targaryen kings, we swiftly get a proliferation of similar-sounding names (Rhaena, Rhaenys, Rhaenyra…), dragons, and characters with named swords. It rolls along enjoyably enough at times, but at other times, such as when Grover, Elmo, Oscar and Kermit Tully are introduced (the latter a young man described as “green as summer grass”), you just get the feeling that you’re being trolled by a multimillionaire author.

Stephen G. Alter, Darwinism and the linguistic image: language, race, and natural theology in the nineteenth century

A little book about the mutual influence of linguistics (comparative philology) and evolutionary biology in the 19th century, convincingly dispelling the myth that it was a one-way street from Darwin to Schleicher (in fact, the opposite is closer to the truth). Should be required reading for anyone who wants to pontificate about language and evolution on the internet. Some of the key figures - Darwin, Max Müller, Lyell, Schleicher - I was familiar with; others (Farrar, Haeckel, Wedgwood) were less well known to me. Still need to read more in this vein. Worth noting that despite the subtitle there’s actually almost nothing about race in here, though a lot about species.

Adam Rutherford, How to argue with a racist

This book will not tell you how to argue with a racist. It will, though, tell you exactly why racists are wrong about basically everything. The book deals with skin colour, ancestry, sporting achievement, and intelligence. If by the end of it you’re not disabused of the idea that genetics can be used to support the idea of differences between races, you’re a moron. That said, the book could have done with better editing at points: it jumps around a bit, and it’s not always obvious why it’s discussing what it’s discussing.

Mary Robinette Kowal, The Calculating Stars

An alternative history of spaceflight, foregrounding women astronauts. I expected better from a book that won the Hugo and Nebula prizes. I’m very much into feminist sci-fi (just check the lists...), but this is wish-fulfilment fantasy with a Mary-Sue protagonist (who has a Gary-Stu husband). Only the treatment of the protagonist’s struggles with anxiety saves this book from mediocrity.

Heinrich Böll, Ansichten eines Clowns

250 pages of a man sitting in his flat, reminiscing and making phone calls, doesn’t sound like a good basis for a book, but it’s a very thought-provoking and rich read. I’ve certainly never read a book that makes Catholicism as an ideology seem so ridiculous. It’s shocking to me that people today still believe this nonsense, and that their lives are restrained and constricted for no reason by ideologies that stem from this.

Steffen Krogh, Die Stellung des Altsächsischen im Rahmen der westgermanischen Sprachen

A study that tries to find out whether Old Saxon forms a closer unit with Old English and Old Frisian on the one hand or Old High German on the other. There are some idiosyncratic moments (such as Krogh’s definition of a “language”) but on the whole this is a careful and intelligent study. Not at all light reading but very clear and sensibly structured. It’s a shame that the conclusion ends up being basically a negative result: we don’t have evidence for one grouping over the other.

David Gaider, Dragon Age: The Stolen Throne

I don’t have high expectations of computer-game spinoff novels, but still I was quite disappointed by this one - it seemed very predictable and clunkily-written. The last third is more interesting because of the character development. Still, hard to avoid the conclusion that the games are better written - odd given that they’re written by the same guy.

Bettelou Los, The rise of the to-infinitive

This book is just excellent - one of a handful that combines corpus linguistics with linguistic theory with really serious philological acumen in the historical study of English syntax. I will certainly find myself coming back to this in the future, as there’s so much detail and insight here.

Babette Ludowici (ed.), Saxones

This is the book that accompanies a recent exhibition at the Hanover and Brunswick state museums about the Saxons in the first millennium CE. It’s beautifully produced and full of interesting information - even if I haven’t learned to love archaeology by reading it. My main takeaway (also see the Flierman book above) is that before the end of the 6th century it’s more or less meaningless to refer to Saxons, at least as the Germanic-speaking inhabitants of northern Germany, since the textual record gives us no support for such an identification. The book also gives interesting perspectives on relations between the Saxons and the Franks - the Saxons were by no means as unChristian pre-Charlemagne as is sometimes made out! - and between the inhabitants of Britain and northern Germany in the fifth and sixth centuries (there’s archaeological evidence for bidirectional trade and movement of people).

Gillian Cross, A map of nowhere

This short book from the author of the Demon Headmaster series is an excellent, unsettling treatment of right and wrong and how they can be identified.

Adam Kay, This is going to hurt

It did. This is about the author’s experiences as a junior doctor in obstetrics and gynaecology, and succeeds in being hilarious and horrifying at the same time - like Scrubs, but with longer shifts and way more genitalia. Having read this I can immediately see why several people I know also decided to leave this career path before it was too late.

Leo Tolstoy, Anna Karenina

This book came highly recommended: lots of people who ought to know what they’re talking about think it’s the best book ever, and it’s hard to think of higher praise than that. Several people I know and respect also have it as their favourite book. Unfortunately, it did not go down well with me. Wealthy Russian woman, married to a man called Alexei, has an affair with another man called Alexei, kind of wants a divorce but also kind of doesn’t, and ends up topping herself. Meanwhile the book’s other protagonist, Levin, whinges his way through the book until right at the end he finds God, basically by voluntarily lobotomizing his higher reasoning skills. And the rest of the 800+ pages are just more wealthy Russians eating caviar, going to the races and talking shit to one another, some of which would presumably have made sense if I’d lived in nineteenth-century Russia. I thought the enduring value of these massive 19th-century tomes was that they shed light on eternal issues related to ‘the human condition’, but there was precious little I could relate to here. If there’s a moral message, I suppose it’s that societal condemnation of divorce is bad, and its treatment is also highly sexist. But I've already got that message, and I live in a world with radically different societal norms, so what was the point? On the plus side, I like Tolstoy’s general prose style: at the level of sentence and paragraph construction it’s breezily easy to read (at least I assume so, and assume that the translation was a good one).

Omer Preminger, Agreement and its Failures

This book makes a powerful case that “derivational time-bombs” - features that lead to ungrammaticality unless something is done in the syntax to sort them out, like the traditional uninterpretable features in Minimalism - can’t be the right way to deal with agreement phenomena like the Person Case Constraint. Rather, we need a theory that allows operations to be triggered but not complete successfully. The main empirical domain is the Agent Focus construction in Kichean, but there is much more here too. Preminger succeeds admirably in showing that something like obligatory operations is necessary - the next question would be whether it is sufficient as a replacement for the classical approach to Agree.

David Gaider, Dragon Age: The Calling

Like the previous book, this one was a letdown - probably more so, as it’s mostly a dungeon crawl. At the end everyone betrays each other, which is entertaining, but I was still left with the impression that Gaider is a better game writer than a novelist.

John Stuart Mill, On Liberty

See my blog post on this one. (Embarrassing editorial note, 12/2020: I never got round to finishing this blog post. Maybe I will some day.)

Giulio Lepschy (ed.), History of linguistics, vol. 1: the eastern traditions of linguistics

A short book compared to others in the series, with no two chapters comparable. Raphael Loewe’s chapter on Jewish linguistics is virtually unreadable, and others (such as the one on Sumerian) reveal that there was no tradition of linguistics to speak of in that culture. One also wonders how they selected the traditions for inclusion: why no Japanese, or Georgian, for instance? Still, there’s a lot of information packed into these 200 pages.

Katerina Somers Wicka, From phonology to syntax: pronominal cliticization in Otfrid’s Evangelienbuch

What this book lacks in breadth of sources (it’s based on a single text) it makes up for in breadth of theory, switching deftly between prosody, grammaticalization theory and syntax. The umlaut facts reported here - applying across grammatical words! - are very cool. I was less convinced by the syntactic bit, but maybe I just like my trees to have more meat (branches?) on them.

Ann Leckie, Ancillary Justice

A cerebral, challenging sci-fi novel with a fascinating protagonist. I found the backstory pretty difficult to follow in the first half, with the result that it was quite difficult to understand what motivated some of the characters in the second half. I’d probably have to reread it to remedy that. But there’s no doubt that aspects of this book will stay with me.

Norvin Richards, Contiguity Theory

I had high hopes for this book, and it’s fair to say I was disappointed. The two main empirical domains - the EPP and wh-movement - are covered by entirely disjoint theories and principles, so that it’s hard to speak of a single “Contiguity Theory”. And though it talks of falsifiability and predictions, the theory construction here is almost completely post hoc: time and again in the book we’re confronted with facts that aren’t covered by the theory so far explored, and another device or auxiliary hypothesis is introduced (“Hippocratic” Altruism; the weird operation of Contiguity-adjunction; etc.). As far as I can tell, the theory also predicts long-distance wh-movement only in non-negative clauses (!). The core theoretical contribution is left unacceptably vague and insufficiently formalized at several points: phonological information must be present in syntax, but what kind of information, and when? Sometimes it seems important that a specific lexical item is phonologically null, other times not, but we’re not given any way to predict which case we’re dealing with. Most importantly, there’s no systematic attempt to contrast the theory presented here, in which phonological information drives syntactic computation, with a theory in which such effects are interface-driven - just a few throwaway comments at the beginning of the conclusion.

Hermann Hesse, Bodensee

A collection of Hesse’s works relating to the Lake Constance area, where he lived with his family from 1904-12. The book was a birthday present from Fernanda, and I had the good fortune to get the boat across to Gaienhofen with Fernanda and Henri while I was reading this book. As a young man in the paradisiacal surroundings of the Bodensee, yet clearly gripped by wanderlust and not entirely content with his lot, the book resonated with me substantially. Also beautifully written, of course.

William Dalrymple, The Anarchy: the relentless rise of the East India Company

I was prepared for this book to be depressing, given that it deals with the corporate looting of an entire subcontinent and the emergence of modern dick-swinging capitalism. It was that, but also clearly and engagingly written - almost an adventure story, except that you’re never rooting for the protagonists. These latter - in particular school-bully-made-good Robert Clive, failer-upwards Charles Cornwallis, and ruthless corporate imperialist Wellesley - are a horrible bunch. This book should be required reading for anyone who believes that the British Empire was a virtuous enterprise. Not that those people are the sort that read books, unfortunately.

Geraldine Heng, The invention of race in the European Middle Ages

This book’s thesis is that race as an essentializing category is invented not in the early modern period but during the Middle Ages. Heng masterfully surveys Jews, Saracens, blackness and whiteness, American Indigenous peoples, Mongols, and the Romani, showing in each case that the characteristics of race-based discrimination are there in all cases. The “show-don’t-tell” style means that sometimes it was difficult for me to follow the argumentation - for instance, in the chapter on the Mongols, when it’s asserted that Nestorian Christians are treated as a race. But there’s no doubt that this is an excellent synthesis and successfully makes the case for its general thesis.

David Gaider, Dragon Age: Asunder

I should have read this before playing Dragon Age 3 - it would have given the character of Cole, who I didn’t much like, a more meaningful role. Good to see what happens to old favourites from the first game, and a bit more thought-provoking than the first two Dragon Age books, but still quite formulaic.

Pearl S. Buck, The Good Earth

A relaxing read in which not a great deal happens, following a Chinese farmer and his family. If there’s any broader message here, I missed it, but it’s a nice story of the different pressures pulling you one way and another in life. I find it very jarring that nearly every spoken sentence begins with “Well, and”, but I guess that was how people spoke in... late 1920s America? And the fate of poor O-Lan is very sad indeed.

Martin Worthington, Teach Yourself: Complete Babylonian

Fantastic overview of the Akkadian/Babylonian language. The verbal morphology to be found in this language is pretty terrifying, but the book makes a valiant effort to break it down into chunks for easier comprehension. It’d be nice if the exercises were a little more varied instead of just “translate this”, but for practical purposes (reading texts) it’s probably the most useful type of task.

Warhammer Battletome: Lumineth Realm-Lords

As a former High Elf player I couldn’t resist picking this up. It’s beautiful and glossy, but as an army it’s likely to disappoint - especially if, like me, you’re not a huge fan of the Alarith “Cow Elf” aesthetic. That leaves you with only six units to choose from, including two special characters and three rather vanilla troop types. Compared to the mighty fifth edition army book there’s not much here. Still, I’d love to give the rules a go.

Nicky Gardner & Susanne Kries, Europe by Rail (2016)

A gift from colleagues on leaving Manchester, I think, and very well placed - it’s bursting with possibilities for rail trips around the continent, mostly of the “slow travel” variety. I was pleased at how many of these routes I’d already touched upon in my travels, but for the areas where I haven’t - Spain, Italy, the Balkans, Russia, Ukraine, Finland - I now have some ideas to play with.

Johanna Flick, Die Entwicklung des Definitartikels im Althochdeutschen: eine kognitiv-linguistische Untersuchung

A well-contextualized, carefully and intelligently conducted corpus study of the emergence of the defnite article in Old High German - primarily of interest to me because of its comparative implications. The key hypothesis around the role of animacy seemed to me to be both somewhat less motivated and less conclusively demonstrated than the text makes out; the use of stats is also not ideal, since the data are crying out for a multivariate approach, and there are no effect sizes. Still, there are useful findings - not least that the definite article is functioning as such quite a lot of the time even in the earliest texts, which ties in with what Sommerer finds for Old English. Would be interesting to do comparable work on Old Saxon.

Patrick Weekes, Dragon Age: The Masked Empire

More enjoyable than its predecessors: a bit of intrigue as well as sword-and-sorcery. Gaspard is much more sympathetic here than he is in the game, despite being opportunistic. I wasn’t always sure what was going on, but it was well-paced and fun.

Ping Chen, Modern Chinese: History and Sociolinguistics

Books in English about Chinese are a rarity, and I learned a lot from this one. Occasionally it verges on the opinionated, and it’s mostly about standardization, including the writing system and its reform, rather than language change in the usual sense. Still, glad I read it.

Peter Godfrey-Smith, Darwinian populations and natural selection

This ambitious book pieces together various insights to put forward a new model of Darwinian evolution - one that is “permissive” (his word, used repeatedly) and prototype-based in allowing for paradigm cases and marginal cases along a number of dimensions. The spirit is inclusive rather than combative, though he does attack the replicator framework for its incitement to “agential” thinking, and has little time for the view that everything should be stated in terms of the gene (gene-based evolution actually comes out as a fairly marginal process, and genes are dubious Darwinian individuals). It closes with a chapter on cultural evolution where he argues that, far from being a specific case of a more general phenomenon of evolution, Darwinian evolution in the cultural domain is one among several processes at play, and only some cultural change benefits from being conceptualized in this way. A rich and interesting book which I’ll have to revisit.

Goethe, Wilhelm Meisters Wanderjahre, Zweiter Teil

Fuck Goethe, with his arbitrary hierarchies and meaningless upper-class bullshit. If the first part was bad, this part was unbearable. I have absolutely no idea why anyone would write a book like this, other than as a form of masturbation. The little platitudes stacked up at the end of books 2 and 3 are about as insightful as those little soundbite quotes one sees on social media superimposed onto an image of a sunset.

Christian Mair, Infinitival complement clauses in English

A well-written, if not thrilling, exploration of non-finite complementation in present-day English using the Survey of English Usage corpus. It’s interesting to see how functionalist work in the 1980s and 1990s positions itself with respect to generative research - in this book reference is frequently made, but at the same time one can see the disconnect that’s developed between the two traditions. For instance, Mair quite rightly points out that Newmeyer’s conception of functionalism in his 1983 book (a direct, synchronic functionalist explanation for everything) is a straw man, and indeed it’s a position that Newmeyer has since rowed back on. On the other hand, Mair argues against transformations by pointing out that the presumed base form is sometimes not attested, and is often not the most frequent/unmarked form. But there’s also no reason why it should be. Anyhow, several things to chew over and revisit here.

Lisa Green, African American English

A nice textbook treatment of AAE - I learned several things. It’s a shame that it has to say “this behaviour is rule-governed, and not just a result of random errors” on virtually every page, but I suppose it doesn’t hurt to hammer home the (socially) most important point.

Luo Guanzhong, Outlaws of the Marsh (The Water Margin, trans. Shapiro)

This 1000-page monstrosity, one of the Four Great Classical Novels of China, is like nothing I’ve ever read. It starts by jumping from character to character as the 108 outlaws find their way to Liangshan Marsh, and gradually morphs into more of a military novel. It’s swimmingly readable, though not always easy to keep track of the protagonists. And the last 100 pages are full of heartbreak as the characters that have been introduced gradually suffer a variety of unfortunate fates.

Frederick Forsyth, Icon

This thriller, in which the Russian tsardom gets reinstated in order to prevent a fascist from attaining power, is bonkers. It didn’t stop me from loving it as a teenager, though, and I enjoyed it a lot this time too. What I didn’t notice the previous time was that it’s awfully misogynistic. And on Wikipedia I found out that Forsyth himself is a Brexiteer, which is unsurprising.

Thomas McFadden, The position of morphological case in the derivation

There are some PhD theses that because of their scope and erudition I can hardly believe are PhD theses, and this is one of them - it blows me away. As far as I’m concerned it’s hard to see any theory of nominal licensing coming back from the absolute brutalizing it receives here.

Liane Merciel, Dragon Age: Last Flight

Last of the five Dragon Age novels, this one has griffons in it, but doesn’t have much else going for it. It reads like good fanfic - not unpleasant to read, but not very page-turny, and not strong in terms of characterization.

Raymond Hickey, Irish English

A rather sprawling book, hugely informative but confusingly structured. Better suited as a reference work than as a cover-to-cover read, I imagine, and I’ll definitely use it as such in future (since I can’t have internalized more than about 10% of the points made).

Esa Itkonen, Causality in Linguistic Theory

This book seems to be an attempt to philosophize an unavoidable way to do linguistics, building on “Autonomous Linguistics” (I think I’ll have to read more about this - it seems to have been developed in the author’s previous work), adding principles of rationality, and insisting on a norm-based approach. It doesn’t succeed in being unavoidable, and my impression is that many of the critiques of other approaches miss their mark; in that sense it’s a hopeful book about what the author would like linguistics to look like (and a somewhat dense one at that).

Max Gladstone, Three Parts Dead

A magnificent urban fantasy novel, very reminiscent of Robert Jackson Bennett’s similarly excellent City of Stairs. Will definitely be after more of this.

David Lightfoot, Born to Parse: How Children Select their Languages

There is both continuity and change here. Continuity in that Lightfoot emphasizes points that can be found as early as his 1979 book, such as the incoherence of an approach that takes “languages” as the primary entities under analysis. Some of the phenomena he investigates are also presented in a very similar way to his previous work (modals and psych verbs in the history of English, for instance). As for change, in this book Lightfoot abandons the Principles & Parameters model of grammatical variation, choosing to replace it with the idea that children discover their grammars by parsing. Unfortunately, nothing at all is said about how this works (other than that the grammar is the parser, following Phillips), so the theory that’s at the core of the book is effectively empty and unfalsifiable. Under this approach, children discover that a verb is in T by... parsing the sentence, and discovering that the verb is in T. I suspect that children would also be able to detect the severe limitations of reasoning like Lightfoot’s in this book. Compounding the issue, Lightfoot repeats earlier, invalid criticisms of the TLA and similar models of acquisition, falsely stating that they involve storage and evaluation of large sets of sentences (see my blog post on this). He also goes out of his way to praise the awful proposals in Emonds & Faarlund 2014 (which, in fairness, are at least empirically evaluable and can be shown to be wanting, unlike the “theory” developed here). It would be nice to have more positive things to say about a book by someone whose thinking, in the past, I’ve always found inspiring and challenging, even when I’ve strongly disagreed with his conclusions. But so it goes.

Baldur’s Gate: Descent into Avernus

A sprawling D&D campaign that doubles as a sourcebook for the city of Baldur’s Gate and for the Nine Hells. I would like to try this out on some players - the content is nice, and there are some fascinating ideas/setpieces, but I’m not so sure how the whole adventure would hang together.

Saturday, December 19, 2020

The TLA does not "store sets of sentences"

In a series of publications, David Lightfoot has made claims about the Triggering Learning Algorithm (TLA) for the acquisition of syntax. The core claim is that the TLA is an input-matching model which evaluates a grammar globally against a set of sentences heard by the child; this is then taken to show that the TLA is not a plausible or attractive model of acquisition. The purpose of this blog post is to show that this core claim is false.

What's the Triggering Learning Algorithm?

The TLA is an algorithm developed by Gibson & Wexler (1994) to characterize the acquisition of syntax by first-language learners. It assumes that the learning task involves setting a finite number of binary parameters: at any given time, the learner's grammar can be represented as just a vector of ones and zeroes. The flowchart below illustrates how it works, informally.

Flowchart for a child hearing a sentence under the TLA

The child receives an input sentence. This isn't just a raw string – rather, it's assumed that the child is able to extract certain syntactic properties unproblematically, so the input sentence, rather than "Mary likes motorbikes", would be something like "S(ubject) V(erb) O(bject)".

The child then checks whether the sentence is compatible with the grammar they have stored in memory. For instance (all else being equal), a sentence "S V O" would be compatible with a grammar where the Head Parameter is set to head-initial and the VP is therefore head-initial, but not with a grammar where it's set to head-final.

If the sentence is compatible with the stored grammar, then nothing more needs be done: the child doesn't change its stored grammar, and will wait until another sentence comes along, at which point the process starts again (the dotted line).

If, on the other hand, the sentence is incompatible with the stored grammar, then the child will entertain an alternative grammar. They do this by picking one of the parameters at random and flipping the value. If (say) there are six parameters, then this basically involves rolling a six-sided dice. For instance, they might flip the Head Parameter from head-final to head-initial. Or they might flip a different parameter – say the Null Subject Parameter, from "boo null subjects are bad" to "yay null subjects".

The child then checks whether the sentence is compatible with the new, alternative grammar. If the sentence is "S V O" and the child has flipped the Head Parameter to head-initial, then the sentence is compatible with the new grammar. This causes the child to replace the grammar they had stored in memory with the new one they're entertaining. If, on the other hand, the child has flipped the Null Subject Parameter to "yay null subjects", that doesn't help them parse the sentence "S V O", and at this point the child gives up (perhaps getting bored) and reverts to the stored grammar. Either way, the child then waits for the next sentence (the dotted line).

After a certain point (after a particular number of sentences, or the end of the Critical Period), the learning process stops, and the child's grammar has reached its final form.

The TLA in this form is pretty straightforward.

Obviously I'm presenting it informally, and glossing over a lot of technical details. In particular, like the exposition in Gibson & Wexler, I'm assuming the Single Value Constraint (the child only tries to flip one parameter at a time) and the Greediness Constraint (the child will only adopt the new, alternative grammar if it actually helps them analyse the sentence they've heard). There's more to be said, but go and read Gibson and Wexler (1994) if you find this sort of thing exciting.

Lightfoot's criticisms of the TLA

Now let's see what Lightfoot has to say about it.
"In being E-language based, these models face huge feasibility problems. One can see those problems emerging with the system of Gibson & Wexler (1994). ... [T]he child needs to determine which ... [grammar] his/her language falls into. That cannot be determined from a single sentence, given the way their system works, but rather the child needs to store the set of sentences experienced and to compare that set with each of the ... possible sets, not a trivial task and one that requires memory banks incorporating the data sets ... So if acquisition proceeds by Gibson & Wexler's TLA and there are forty parameters, then there will be over a trillion different data sets to be stored and checked, and each of those data sets will be enormous." (Lightfoot 2006: 76)
From the characterization of the TLA provided in this blog post, we can see immediately that the statement that "the child needs to store the set of sentences experienced" is incorrect. All the child ever needs to store long-term is the grammar, and short-term a single sentence needs to be held in working memory. Once a single sentence has been dealt with, it can be immediately forgotten. Nothing about the TLA involves storage of sets of sentences, or comparison of sets of sentences.

The same criticism is repeated in subsequent work by Lightfoot:
"Work in synchronic syntax has rarely linked grammatical properties to particular triggering effects, in part because practitioners often resort to a model of language acquisition that is flawed ... I refer to a model that sees children as evaluating grammars against sets of sentences and structures, matching input and evaluating grammars in terms of their overall success in generating the input data most economically, e.g. ... Gibson & Wexler (1994), and many others." (Lightfoot 2017a: 383)
There is no overall evaluation of the success of grammars in generating the input data under the TLA, as shown. There is also no role for economy.
"Gibson & Wexler['s] ... children are "error-driven" and determine whether a grammar will generate everything that has been heard, whether the grammar matches the input. ... There are huge feasibility problems for ... these global, evaluation-based, input-matching approaches to language acquisition, evaluating whole grammars against comprehensive sets of sentences experienced (Lightfoot 1999, 2006: 76f). ... [I]n order to check whether the generative capacity of a grammar matches what the child has heard, s/he will need to retain a memory in some fashion of everything that has been heard." (Lightfoot 2017b: 6–7)
As clarified above, the child does not need to remember everything that has been heard, under the TLA. In fact, the child does not need to remember anything that has been heard, in the long-term. There is no global evaluation.

Lightfoot pushes these points particularly strongly in his latest book, Born to Parse (2020). Here he states that systems such as the TLA
"... involve the global evaluation of grammars as wholes, where the grammar as a whole is graded for its efficiency (Yang 2002). Children evaluate postulated grammars as wholes against the whole corpus of PLD that they encounter, checking which grammars generate which data. This is "input matching" (Lightfoot 1999) and raises questions about how memory can store and make available to children at one time everything that has been heard over a period of a few years." (Lightfoot 2020: 24)
These questions do not arise, because the TLA does not need to store input. There is no evaluation of grammars as wholes.

(It's also not clear why Yang 2002 is cited here, because his model of acquisition also doesn't involve global grading of the efficiency of grammars.)
"Gibson and Wexler take a different approach, but in their view as well, children effectively evaluate whole grammars against whole sets of sentences ... [W]hen they encounter a sentence that their current grammar cannot generate ... children ... pick another parameter setting, and they continue until they converge on a grammar for which there are no unparsable PLD and no errors. ... Gibson and Wexler's child calculate[s] the degree to which the generative capacity of the grammar under study conforms to what they have heard." (Lightfoot 2020: 25)
The TLA does not require that the child converge on a grammar for which there are no unparsable PLD and no errors. In fact, when the child updates their stored grammar, the result may be to make sentences that were previously heard (and parsed correctly) unparsable in future. The child does not calculate the degree of anything: it's a fully discrete model of parameter setting, which doesn't even involve stored probabilities, unlike subsequent models such as that of Yang (2002).

In Lightfoot (2020), these (invalid) criticisms of the TLA and models like it are referred to constantly in subsequent chapters (e.g. p35, p60, p64, p91) in order to motivate a different approach, one which it would take us too far afield to discuss here.

So has the TLA been saved, then?

My goal in this blog post is not to defend the TLA as a viable theory of acquisition per se. There are lots of actually or conceivably valid criticisms of it. For instance:
  • For certain systems of parameters, local maxima or 'sinks' occur – grammars which it is impossible for the learner to escape from (Gibson & Wexler 1994; Berwick & Niyogi 1996).
  • Learning is more efficient if the Single Value constraint is dropped – that is, if the child is allowed to flip any number of parameters at once and jump to a completely different grammar (Berwick & Niyogi 1996; Niyogi & Berwick 1996).
  • The system of parameters explored by Gibson & Wexler (1994) can be learned by a much more conservative learner who just waits for the right unambiguous 'silver bullet' sentences to come along (Fodor 1998).
  • The TLA is very vulnerable to noisy input data: if the very last sentence the learner hears before fixing their grammar once and for all happens to be junk, the learner may flip to a different grammar, and "the learning experience during the entire period of language acquisition is wasted" (Yang 2002: 20).
  • The TLA falsely predicts abrupt changes in the child's linguistic behaviour, and cannot capture gradualness or probabilistic variation (Yang 2002: 20–22).
  • The TLA is dependent on a finite set of innate parameters in order to function; if this turns out not to be how syntactic variation works, then the TLA won't work either.
Perhaps because of some of these problems, the TLA has effectively been abandoned as a model of syntactic acquisition over the last two decades: at least as far as I am aware, no one is seriously working with it as a real contender (rather than as a toy model for expository purposes) at the moment. It's therefore not obvious why Lightfoot spends so much time attacking it. But there's no need for speculation on that front, because, as I've shown in this blog post, his criticisms of the TLA are invalid, and depend on imputing properties to the model that it does not actually have.

References

  • Berwick, Robert C., & Partha Niyogi. 1996. Learning from triggers. Linguistic Inquiry 27, 605–622.
  • Fodor, Janet D. 1998. Unambiguous triggers. Linguistic Inquiry 29, 1–36.
  • Gibson, Edward, & Kenneth Wexler. 1994. Triggers. Linguistic Inquiry 25, 407–454.
  • Lightfoot, David W. 1999. The development of language: acquisition, change, and evolution. Oxford: Blackwell.
  • Lightfoot, David W. 2006. How new languages emerge. Cambridge: Cambridge University Press.
  • Lightfoot, David W. 2017a. Acquisition and learnability. In Adam Ledgeway & Ian Roberts (eds.), The Cambridge handbook of historical syntax, 381–400.
  • Lightfoot, David W. 2017b. Discovering new variable properties without parameters. Linguistic Analysis 41, 1–36.
  • Lightfoot, David W. 2020. Born to parse: how children select their languages. Cambridge, MA: MIT Press.
  • Niyogi, Partha, & Robert C. Berwick. 1996. A language learning model for finite parameter spaces. Cognition 61, 161–193.