'totally novel sentence'

[Dan I. SLOBIN]

This was never a puzzle to me; rather, I found it a great insight when I
first encountered it as an undergraduate in the late fifties, and have
been passing it on to my students ever since.  Simply put:  you almost
never enocunter the same sentence twice--i.e., syntactic construction plus
lexical items.  (What is novel is each unique combination of words and
morphosyntactic patterns.)  The consequence is that, although you can learn
your vocabulary by rote, you can't learn your sentences by rote.  Ergo,
language acquisition must be conceived of as a "generative" or
"constructivist" accomplishment.  The achievements of recent decades of
linguistics do not dim this insight, but merely make it more important:
to be sure, the child learns syntactic patterns, constructions, rich
lexical entries, and so forth--but each actual utterance, produced or
received, calls upon general processing skills (to use another old term,
"competence").

-Dan Slobin
Psychology, UC Berkeley

On Tue, 23 Jun 1998, Esa wrote:

> Dear colleagues
>         I am about to finish a paper on which I have been working some
> time and, just to make sure that I have got everything right, I have to
> ask you the following question. In the not-so-distant past it was widely
> claimed that speaker-hearers constantly encounter and understand
> (utterances of) 'totally novel sentences'. Did anyone of you understand
> what was meant by this curious statement? I certainly did not. If I know
> the language in question, every sentence that I hear has some obvious
> similarities (or analogies) to sentences that I have heard before. I
> never hear sentences exemplifying totally novel sentence structures or
> containing totally novel grammatical morphemes, and I seldom hear
> sentences containing totally novel lexical units. So, to repeat, did
> anyone of you ever understand what was meant by this often-repeated slogan?
>
> Esa Itkonen
>