Here’s a “cookbook” of various kinds of English syntax trees. This is my attempt at capturing a happy medium of what’s commonly introduced in introductory texts and my own undergrad syntax class.
Perhaps the most controversial assumption is that morphology is a product of syntax. Not taking sides here - all I’m saying is this is what I was first taught.
- Headedness principle
- A phrase needs a head; an XP needs an X.
- Binarity principle
- Everything has two children (which may be empty or host a combination of multiple things due to movement).
- The presence of an X necessitates the presence of X' and XP above it.
- VISH (Verb-Internal Subject Hypothesis)
- The hypothesis that subjects of verbs generate in SpecVP and raise to SpecTP.
- Head-Movement Constraint
- Heads can only move to adjacent heads.
You can have as many nested X’ levels as you want. The specifier position of XP is abbreviated SpecXP.
The “subject” generates in the specifier of a lexical phrase before raising to the specifier of a functional phrase, much like SpecVP raising to SpecTP per VISH.
The subject generates in SpecVP and raises to SpecTP as per VISH. T lowers to V since [present] is not pronounceable on its own; √eat+[present] resolves at a later stage to “eat”.
Things that can occupy T are features [past] [present] and modals.
Question with a modal
Questions generate with a phonologically null [question] C that need to combine with something to be pronounced; here, the T raises to C.
Question with do-support
As in the question with a modal, T raises to C, but as the [question] C and [present] T are both unpronounced, the repair strategy is to resolve the combination to “do”.
Negation with do-support
T-to-V movement, like in the “classic” SVO sentence, is blocked due to the Head-Movement Constraint. The repair strategy is to resolve the unpronounceable [present] to “do”.
Question with aspect be/have
Last updated January 24, 2024