IONCE HEARD A lecturer defend the proposition that "There's no such thing as bad dance," and she made a fairly compelling case. The point was that dancing, on a professional level, makes such excruciating demands on brain and physique that it can be done completely or not all. In classical ballet, for example, a man can either do a double turn in the air or he cannot; a woman can do multiple pirouettes or she can't - there are no intermediate degrees.

Choreography is another matter. There's lots of bad choreography. It isn't only the difficulty of preserving dance that has left us with such a small legacy of past choreography, but that so little of it seemed worth saving.

The qualities that make choreography bad are no different than those than stigmatize bad music or poetry or film - pretension, unjustified long-windedness, obscurity, shapelessness, pointless repetition, cuteness, bald imitation, outright kitsch. If you recall the Bolshoi Ballet's historic-didactic monstrosity called "Spartacus," then you are acquainted with almost all of these blemishes in one mammoth package, with the possible exception of obscurity and cuteness. "Spartacus," intended as an homage to the spirit of independence of the Russian people, came out looking like nothing so much as one of Hitler's fanatically regimented rallies - square, blunt, brutal and flagrantly militaristic.

Few ballets are as exquisitely, purely bad as this, but clinkers are rarely in short supply: Balanchine's "Who Cares?," which tries to render Gershwin's jazzy nonchalance in terms of toe-shoe ballet, and succeeds only in underscoring the incongruity; John Taras' insipidly "modernized" version of "Daphnis and Chloe," which might be subtitled "Hell's Angels on Mount Olympus"; Gerald Arpino's "Confetti," so frantic in its pursuit of synthetic gaiety tht it hurts to watch; and Norbert Vesak's "What To Do 'Til the Messiah Comes," a title that suggests the answer - stay away from this oafishly trendy ballets that are not so much bad, but rather not as good as a passing fad makes them out to be - Twyla Tharp's "Push Comes to Shove," for instance, which is still enjoying a popular vogue, but which really stands up very poorly under careful, repeated scrutiny. Relative newcomers to ballet like it because watching a famous dancer do hat tricks makes them feel more comfortable with an art so commonly thought of as precious and fey. The combination of Haydn's sober classicism and the jazzy, quirky ties that are Tharp's movement specialty has the same effect.

But the joke quickly wears thin. and there's not enough solidity underneath to sustain the conceit over four movements. Tharp is an original and often brilliant choreographer, but "Push," far from being her "masterpiece," is her concession to mass cult. It reminds me of Jascha Heifetz performing Gershwin on the fiddle - he could play the notes superbly, but it seemed a pointless waste for him to try.