Book review 1: He is a Linguist

I have started reading “I am a Linguist,” and I’ve decided to “review it” here in sections (as I’m unlikely to get the chance to read it all in the near future). So, bear in mind that in writing this, I haven’t read the whole book.

Some initial disclosures and potential conflicts of interest: I was an undergraduate student of the author of this autobiography when we were both at the Australian National University. Nowadays I admit to some pretty extensive differences of opinion in approaches to fieldwork, student supervision, in what constitutes appropriate interaction with one’s colleagues, and in most aspects of Australian prehistory, both the details and the big picture. I am therefore not exactly a neutral reader of this book. (I should also say at the outset that the closest I am ever likely to come to writing an autobiography is the Bardi grammar.) I also have a little reticence in dissecting a former teacher online and in public, and I’d ask readers to keep that in mind if they feel moved to comment.

In reading, I have tried to approach this book on its own terms, as an example of its genre (for which see below), and as though I have no additional information about the author other than what he presents. It strikes me that this is an excellent book for critique of the idea of “the text”. Autobiographies are highly constructed; a chance for the authors to reflect on their own lives, to present a position of themselves in the world, to explain or defend their actions, to give their take on events that they were involved in, or to shame and praise their acquaintances. I don’t know which this book is going to be yet; at the time of writing our young hero has just taken his O-levels.

The book begins with a ‘day in the life of a field linguist’, or more specifically, a field linguist in a remote village in Amazonia. I imagine that many field workers who were asked to describe a day in their field lives would construct a chapter somewhat along these lines, and the author gives quite a good sense of the mixture of excitement and boredom, of vague irritation and great community feeling, of the difficulties of getting regular exercise or doing purposeless wandering, and the like. Some might have more self-doubt, some might concentrate more on community intrigues, or on relationships between the linguist and other community outsiders, but these are all things that differ from linguist to linguist and community to community. Coming through in this first chapter I recognised the infectious enthusiasm for linguistics that many of us trained at ANU in the 70s, 80s and 90s would know.

The first chapter, it seems to me, is written as pedagogical. It’s the sort of thing that I might assign to my fieldwork students (or, actually, more likely, to freshperson seminar students) to give them a taste of one area of linguistics. One reason I think the chapter is constructed for an audience like that (or the general public), and not for linguists, is the sample ‘problems’ that the author puzzles over in the chapter. They are facile for anyone with experience of exploratory data gathering, not the sort of things that one would puzzle over for hours or days in the field, but exactly the sort of thing that one would use in an example radio broadcast or 101 lecture.

Having started on this attempt at audience identification, I proceed to the next chapter (on early years at school and family), and now I’m faced with a problem. Given that in the previous chapter I’ve deduced various things about the intended audience, and why particular examples are used, I now don’t know how far I can do that in other chapters. With every anecdote, I keep wondering – why are we getting this piece of information? What picture is the author trying to present of himself? What is he signifying by this admission? How do we constrain (or delimit) the intertextuality that’s invited by this genre? I guess normally we wait until the author’s no longer with us.

One that puzzles me is the emphasis in the book so far on a) the author’s difficulty with learning languages, and Latin in particular (related to a difficulty and/or dislike of memorisation of minutiae); and b) the emphasis on rapid grasp of mathematics and the ‘big picture’. So many Australianists started out as Classicists that we do form a bit of a club, with a shared approach that comes from that tradition; perhaps there is an attempt at differentiation here. In respect of the second point, I found the anecdote of a staff meeting with our author as HoD and calculating grades faster than the rest of the staff were doing on their laptops a little unlikely; did ANU staff really have laptops before 1990?

One other thing I found a little puzzling was the shift from the first chapter of a ‘day in the life’ to the early years in chapter 2. One thing that struck me was how though there is some social analysis in the first chapter, this is altogether absent from the second. Social categories and behaviours at the school are discussed from an adult perspective, but are accepted as ‘fact’ rather than analysed ethnographically (unlike life in the village). One is left with the impression that ethnography is what we do to other cultures, not our own (not that I lay any claim to membership of the community of practice that comprises 1950s English public schoolboys, and since I consider such a group an ideal source of ethnographic analysis, perhaps I am simply confirming the very point I am arguing against).


5 responses to “Book review 1: He is a Linguist

  1. No laptops; some of us had calculators, I had one that was somewhat programmable, and he definitely was sometimes faster than me using it.

  2. The book’s preface ends ‘And thanks to Norbert Wiener for putting the idea into my head, half-a-century ago, with his memoir I am a mathematician.’ (p.xiii) While I find Wiener’s book more revealing of its author and less posturing (and Wiener has no use for code names), there are noticeable parallels between the two: both in what is chronicled (such as the shaping role of the father), and in the expository style: accounts of events and anecdotes mixed with many opinons and with popular overviews of technical aspects. And how about this:

    “… Wiener would occasionally take time out from lecture to describe the plot of his latest pseudonymous detective novel” The legacy of Norbert Wiener: a centennial symposium in honor of the 100th anniversary of Norbert Wiener’s birth, p.17

  3. Interesting what you say about an “emphasis on rapid grasp of mathematics” when in fact his mathematical intuition is so poor — I’m thinking in particular of his “proof” that contiguous languages borrowing from each other, given sufficient time and no invention of new words, will end up with 50% shared vocabulary. (Actually, of course, given his assumptions, the maths shows that they end up identical!) He first came up with this in 1972 in his first book. No big deal, I thought, surely someone will sort him out. But it’s still there in his 2003 Green book, a whole section about it, at the end of which he references Alpher and Nash (1999), where, if I remember correctly, they did do the maths correctly and pointed out, very gently, that he’s wrong.

  4. That 50% equilibrium figure has an interesting history. It’s a ‘thought experiment’ in the 1972 Dyirbal grammar, with plenty of hedges and and no claim that it’s a full model. By the 1997 book, it has become “Australian languages have an interesting property…”. The 2003 reference to Alpher and Nash is something like “Alpher and Nash (1999) provide further discussion,” which seems to me to be a bit of a fudge, since as you point out, they show a) that it’s wrong, b) that in their sample, there aren’t any languages with 50% common vocabulary, and c) if there is an equilibrium, it would be much lower.

  5. “… and c) if there is an equilibrium, it would be much lower.”

    More precisely, the equilibrium common vocab can be anything between 0 and 1. It’s a function of the proportion of new words that are borrowed (L in Alpha and Nash). The maths gives
    p(eqm) = [1-sqrt(1-L*L)]/L
    This assumes that each language has n=2 neighbours. For n>2, p(eqm) would be a bit less.

    For L=1 (Dixon’s assumption), p(eqm)=1 (complete merger)
    For p(eqm)=0.5 (Dixon’s claim), L would have to be 0.80 (20% of new words not borrowed)
    For L=0.37, as estimated for the SW Cape York languages by A&N, the formula gives p(eqm)=0.16. — which is in fair agreement with what’s found in practise.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s