We have seen that words can go together to form constituents.
Words don't just go together in a random, structure-less lump. Instead, when you combine words, different words may be playing different roles in the constituent. There are three types of roles that words in a constituent can play: head, complement, or adjunct. (Some theories of syntax include more roles than this—for example, X-bar theory and Minimalism include "specifiers" as well. However, the roles of head, complement, and adjunct are roles that exist in any theory of syntax, whereas other roles like "specifier" are concepts that only exist in certain theories and not in others.)
The head of a constituent is the word that determines that constituent's functional category. For example, "ball" is a Noun Phrase, and "big stupid red ball" is also a noun phrase; thus, "ball" is the head of "big stupid red ball". But the constituent "kicked the big stupid red ball", this constituent is now a verb phrase, which means the head of that constituent is "kicked".
One of the basic ways of forming a constituent is to combine a head with a complement. A complement (also called an object) "completes" a head; it's a word that a head needs to have, in order to become a complete phrase.
For example, the verb put needs to have a thing that is put, and a place where the thing is put; it needs two complements. If the verb has both complements (sentence #1 below), it can make a grammatical sentence. If it's missing either complement (sentences #2 and #3), or both complements (sentence #4), it will be ungrammatical.
Each head has a specific number of slots for complements. For example, 吃 ("eat") can take one complement; if you try to give it two complements, you'll get an ungrammatical sentence:
On the other hand, 喂 "feed" (like put) is allowed to have two complements (a noun and a verb phrase):
You may notice a challenge to what I've said so far. So far I said that a head has a specific number of available slots for complements and it cannot have too many or too few complements. In reality, though, there are some heads that are flexible (e.g., in English, I can say "I read a book" or just "I read"). These might be 'optional' complement slots, or these two versions of "read" might be slightly different words with slightly different rules. Different theories have different ways of handling this issue.
The last type of role is that of an adjunct. An adjunct, like a complement, is something you combine with a head. The difference is that complements are necessary, and each head has a limited (and fixed) number of slots for complements; we saw examples of that above. Adjuncts, on the other hand, are not necessary, neither are they limited. They just give extra information. In the sentence below, yesterday is an adjunct, because I could take it away without making the sentence ungrammatical (unlike stuff like "on the table" from the previous examples):
Likewise, I can add as many adjuncts as I want in the sentence; it might start to sound weird and hard to remember, but it will never become ungrammatical. This is different from the examples of complements above, where adding an extra complement would make the sentence ungrammatical. With adjuncts, we can just keep adding and adding:
Ultimately, all constituents (and all sentences) are made by combining heads, complements, and adjuncts one at a time. Think of a sentence like The kid rode the bus. We can put this together by combining two constituents at a time, over and over again:
(In phrases like "the bus", the is generally considered to be the head, and bus the complement, and thus we call these Determiner Phrases [the is a determiner] rather than Noun Phrases. The reasons for that are beyond the scope of this subject, but you'll learn about it if you ever take a syntax class. For a phrase like rode the bus, you can use the above concepts to figure out on your own what's the head and what's the complement.)
Linguists often illustrate these with tree structures like the one below; the picture below shows the same information as what I just showed in the steps above.
![Syntax tree for [ [the kid] [rode [the bus] ]](syntaxtree.png)
These trees are also useful for identifying constituents. A constituent is any chunk that you see together in the tree, without any other stuff attached to it. For example, "rode the bus" appears in the tree, and you can do constituency tests to prove that it's a constituent (try this yourself). On the other hand, "kid rode the" doesn't appear on its own in the tree, and you can also do constituency tests and see that it's not a constituent (try this yourself).
You won't need to draw trees in this class, but you may find them helpful.
Continue to the activities below to practice these concepts.
Choose any one of the ambiguous sentences from the previous activity. Focus on just the ambiguous part (you don't need to analyze the entire sentence). For each interpretation, explain how the words went together (and in what order), and identify the heads, complements, and adjuncts. You may do this with a tree or with a step-by-step explanation.
Reflection time! This module just focused on some basic concepts and we have not yet gotten to analyzing specific features of English, Chinese, or other languages; that will be done in the advanced syntax modules, but the basic concepts from this module will help you with those.
First, though, write a reflection about this module (following the same guidelines as the reflection portions of the previous modules).
When you finish this activity, you are done with the module (assuming all your work on this and the previous tasks has been satisfactory). You can return to the module homepage to review this module, or return to the class homepage to select a different module or assignment to do now.
by Stephen Politzer-Ahles. Last modified on 2021-04-23. CC-BY-4.0.