myself with (a) how the thought got there, nor with (b) what events in the f-mind trigger the language faculty into
expressing it.) The initial eventhas to be a call to the lexicon: what words potentially express parts of this thought? As
in perception, it is an open question whether the call to the lexicon should be thought of as an actual“call”from
working memory or as the lexicon actively intruding itself on working memory.
Following Levelt and many others, it appears that the call to the lexicon results in the activationof a variety of lexical
items of varying degrees of appropriateness for the part of the thought in question. In present terms, this can be
thought of as binding the candidate items to working memory, where their conceptual structures compete for
integration intothe thought in conceptual workingmemory. At some point a best candidate wins the competition (the
competition is resolved), based among other things on its match to the intended thought. This last step is standardly
called“lexical selection”in the production literature.^102
Two kinds of speech error result fro mincorrect lexical resolution/selection. If for so me reason the wrong co mpetitor
is chosen, we get word substitutions based on similarity in meaning, sayHe shot him through with a swordfor...arrow.If
resolution/selectiongets fouled up and twowords remain in competition, we get blends such astroblemfortroubleplus
problem. (The mechanisms for the latter as we pass through to the phonology are not so clear, though.)
Oncea lexicalitem's conceptual structureisbound toworkingmemory, itssyntacticand phonologicalstructures ought
to follow suit shortly and bind to their respective“blackboards”—without yet being integrated into structures. It is a
pointof logicthat phonologicalintegration has toawaitthecompletion of syntacticintegration. The syntax determines
what order words come in, which determines the order of phonological integration (remember, this is all being done
incrementally: the processor doesn't have tofinish the syntax or even the word choice for the whole sentence before
starting to integrate the phonology).
In addition to linear order, the phonology must also know what morphological forms of words to choose, depending
on the syntactic context. For instance, it has to know whether a verb is, say, in past tense or ste mfor m, as inHe went
versusDid he go?—which is determined by syntactic context. It also has to know whether to put in morphemes that
lack semantics, for instanceIt's raining, He didn't go, a picture of Bill, and, in languages such as German and
212 ARCHITECTURAL FOUNDATIONS
(^102) It appears therefore that different subprocesses are called“selection”in the perception and production literatures: they correspond to what are called here“binding”and
“resolution”respectively.