Lecture notes for Class of February 1, 2010
LIN100Y (Diane Massam)
Note: these notes focus on material that is not in the textbook.
BE SURE TO READ CHAPTERS 5 and 6.
Semantics: the study of meaning. A central question is where do we place Semantics in
our generative grammar? (which so far, looks like the following):
It is clear that we need some meaning in the Lexicon. This consists of the meaning of
lexical items, which is very hard to represent (see discussion in the textbook about the
meaning of colour terms like ‘blue’, and other aspects of lexical meaning such as
antonymy etc.). We will focus today on one aspect of lexical meaning, and explore how it
is realized in syntax, namely the thematic argument structure of words, focusing mainly
FIRST: Let’s look at how meaning can be reflected in structure, in terms of
modification. The textbook places modifiers as daughters of XP and sisters of X’ (p.
[Remember how when we talked about complements (p. 158-9 textbook) I mentioned
that the book doesn’t really distinguish complements from modifiers. I said we really
have to do this in a complete grammar – For example, on p. 158, Table 5.5 the text talks
about “with a crowbar” as a complement (sister of V), but on p. 210 it talks about “with
binoculars” as a modifier (sister of V’). Be aware that the two are different, but don’t
worry too much about being able to tell for a given PP, whether it is an argument or a
mofifier. Generally complements elaborate on and express part of the meaning of a head,
whereas modifiers add extra meaning.]
Modifiers are part of the phrase they modify, thus structure reflects meaning. Some
sentences or phrases are structurally ambiguous, because there are two structures that
correspond to one string. E.g. “Nicole saw people with binoculars” – the PP can be
modifying the NP or the VP. See trees and discussion for this and coordination on p. 209-
NOW: Back to thematic (also called “theta”) argument structure. Read about theta roles
on p. 210-214.
In languages of the world, if there is no special morphology present, the subject will
express the agent of an active transitive clause (like ‘eat’ ‘kill’ ‘hit’ etc.) and the object
will express the theme (also called the patient sometimes). Note that the second sentence,
the Passive, where this is not the case, includes extra morphology (the verb ‘be’ and the
‘–en’ ending on the verb).
a) The thief stole the painting.
b) The painting was stolen (by the thief).
Pragmatics, or knowledge of the world, helps us know who is the do-er and the done-to
here, but this is not the case below, yet it is equally clear in these sentences who is the
agent and who is the theme.
c) Mary hit Susan.
d) Susan was hit (by Mary).
Linguists have thus posited that there is a universal mapping of AGENT to subject, and
THEME to object position at D-structure. This is supported by the fact that verbs show
closer relations with their themes than with their agents. For example, there are many
verb-theme idioms but no verb-agent idioms (e.g. kick the bucket, throw a fit, etc.)
Thus, there is a mapping from the lexical thematic roles of the verb, to D-structure. Early
Generative grammarians argued that all meaning was present at D-structure, and so
semantics was computed in the lexicon and at D-structure, but no-where else. For
example, in “What did Mary eat”, we know “what “ is the theme because it was merged
as the D-structure object, and even though it moves, this doesn’t change because its
meaning was determined at D-structure. NPs keep the theta roles they got at D-structure.
What about the Passive sentences, as in (b) and (d)? Their function is to downplay the
agent and highlight the theme. They involve “NP Movement”. This is a third type of
movement, in addition to Head Movement and Wh Movement we have already seen. A
derivation for Passive is on p. 179-80 of the text book.
We will not focus on the “by-phrase”, but I assume it is a VP modifier, so hangs from VP
as a sister to V’, like other VP modifiers. The book says it is a complement (p. 179) –
This is an instance of the issue discussed above about the lack of clarity about the
difference between complements and modifiers. You can put it in either place.
Lecture notes for class of february 1, 2010. Note: these notes focus on material that is not in the textbook. Be sure to read chapters 5 and 6. A central question is where do we place semantics in our generative grammar? (which so far, looks like the following): It is clear that we need some meaning in the lexicon. This consists of the meaning of lexical items, which is very hard to represent (see discussion in the textbook about the meaning of colour terms like blue", and other aspects of lexical meaning such as antonymy etc. ). We will focus today on one aspect of lexical meaning, and explore how it is realized in syntax, namely the thematic argument structure of words, focusing mainly on verbs. First: let"s look at how meaning can be reflected in structure, in terms of modification. The textbook places modifiers as daughters of xp and sisters of x" (p.