At some point I shall have more to say about the ‘New Criticism’ of the 1940s and its successors since then – the various ill-starred attempts to remove the subjective from literary criticism and thereby gank some of the prestige (read: grant money) hitherto reserved for the hard sciences. This program, as I have mentioned before, led to the ludicrous practice of analysing literature without any reference either to the intentions of the writer or the reactions of the reader; as if the mere text were an eternal and uncreated thing, existing solely to be studied in the abstract, and not a dirty, low-down, wilful attempt at communication.

Linguistics, which (almost alone among the social sciences) ought to be a science, and can at least be approached as one, is in a worse state than all the others. So I found out a decade ago, when I made the mistake of paying tuition to study it. The ‘Quantitative Methods’ in that field, as in most of the social sciences, consist chiefly of misapplied statistics and a smattering of logic. But if language is anything, it is an attempt to transmit a signal successfully; and you cannot really understand how signals work without studying information theory. Naturally, there was no mention of information theory in the linguistics syllabus; probably because the linguists don’t know any information theory themselves, and don’t even know that it’s there not to know. (These are the same people, in some cases, who laughed at Donald Rumsfeld’s ‘unknown unknowns’; more fools they.) You see, information theory is taught by the Maths Department, and requires other mathematics courses as prerequisites; at the university I attended, it was a third-year course, and by that time a linguist is supposed to have completed all his Quantitative Methods courses and relapsed into comfortable innumeracy.

The inimitable Tom Lehrer, in one of his lesser known songs, took a shot at the same tendency in the social sciences. In his younger days, the social sciences were (as he puts it) desperately trying to justify the word ‘science’ in their title. Social scientists, whose ostensible subject was the study of the nature and interactions of human beings, were instead abandoning that subject to go in for the aforementioned Quantitative Methods. It was this absurdity that spilled over into literature; and pretty nearly everything that needs to be said about it was said briefly and pithily by Mr. Lehrer in the song that follows.


Lyrics by Tom Lehrer
Music from ‘Choreography’, by Irving Berlin

Is the change
They’re trying to arrange
Today in sociology.

In their attics
Are learning mathematics
Just for sociology.

By equation:
They all feel it’s much more satisfactory.
They, in an ivory steeple,
Far away from all people,
They do research in sociology.

Who wrote lies
Now present them in disguise—
A cinch in sociology.

A tract,
Quite abstract,
Without one single fact
Is splendid sociology.

Who used words
Now all talk in terms of X and Y and Z.
They can take one small matrix
And really do great tricks
All in the name of sociology.

Who wrote prose
Now write algebra, who knows?
It may be sociology.

Full of sigma and chi-square
And full of s—ociology.

They consult,
Sounding occult,
Talking like a Mathematics Ph.D.
They can snow all their clients
By calling it science,
Although it’s only sociology!


  1. How else can people who can’t stand math going to get degrees?

  2. antares says:

    I almost made the mistake of seeking an advanced degree in Linguistics. Then I found out that the field’s great light was Noam Chomsky. That put an end to that nonsense.

    The crippling handicap of Linguistics is the same as Economics. The practitioners seek to prescribe before they describe. They fail to understand how things are. Instead they tell how things should be.

    I admit that there are a few who grind away in obscurity and add to the sum of human knowledge. In time, only their work will remain.

    • There are indeed a few grinding away: like the linguist from India whom I first heard of a year or so ago, who constructed a rigorous proof that Chomsky’s ‘universal grammar’ (or ‘generative grammar’, or ‘runcible grammar’ or ‘thingumbob grammar’; he changes the name every decade or so) is an impossible project, because, in essence, Chomsky is trying to use computational methods to solve a non-computable problem.

      I had a pretty strong intuition that this was the case, back when I was a student in ’07; but I had not the formal logical chops to construct a rigorous proof, and I was tickled pink to see that this chap had done it. —And so much for Chomsky, whose sole enduring contribution to linguistics has been to make generations of linguists ignore actual languages and go haring off after the Philosopher’s Stone of the universal-generative-super-dooper-ultimate-X grammar.

  3. Hah! It’s a parody of Irving Berlin’s Choreography sung by Danny Kaye in White Christmas! An excellent match with the mocking tone of the original.

  4. As a hard scientist – yup, a physicist with a PhD in Nuclear Engineering/plasma physics – I object to this attempt to apply the term ‘science’ to things that are so clearly (to us, anyway) not.

    We shouldn’t have had to add ‘hard’ to science just to get away from the ‘soft’ and ‘squishy’ versions.

    I had health problem recently which required me to read and understand around a hundred research papers in medicine – what a mess! As a result, I don’t believe ANY of them (well, except for the nice paper with the WRONG title which summed up everything I had read until that point, and would have saved me a LOT of work).

    How to LIE with STATISTICS and LINGUISTICS was invented for these researchers.

Speak Your Mind