We found in the previous section that by taking into account the subclassification information of the main category, we can express more generalizations about the properties of verbs. Another characteristic of this type is the following: expressions of category V are headings of sentences of category VP. In the same way, the heads of Nps, as (i.e. adjectives) are heads of PPS and ps (i.e. prepositions) are heads of PPs. Not all sentences have heads – for example, it is by default to say that coordinate sentences (for example. B the book and the bell) have no heads – yet we want our grammatical formalism to express the parent-child relationship where it applies. At the moment, V and VP are just atomic symbols and we have to find a way to relate them using functions (as we did earlier to relate IV and TV). Compared to English, Latin is an example of a very volatile language. The consequences for the agreement are therefore as follows: 2In addition to the literature debated here, there is important work on what is known as “multiple final marking” in languages such as Greek (see Alexiadou 2014 for an overview). However, the agreement to finalize MSA is very different from the definitive mark repeated in Greek, which has a semantic effect and can only appear on predictive adjectives, which supports an analysis of the reduced relative content of the phenomenon. In contrast, MSA`s confirmation agreement is semantically empty and consistent for all attributive adjectives, regardless of their predicativity.
These differences indicate that the two phenomena are not the same and also support a PF-based analysis on the MSA determination agreement under a Y-based grammar model. The restriction that DP AGR must dominate in order for feature copying to take place predicts the adequacy of certainty for attributive adjectives and not for predictive ones. Given the basic DP structure that contains an attributive adjective in (13/17), the aP DP dominates, which makes it possible to copy the functionality. However, in the predictive structure (14/16), the DP subject does not dominate aP. The DP determination value cannot be copied to the AGR node of the predictive adjective. This then explains the consistency of coherence between an attributive adjective and its noun and the absence of certainty agreement on predictive adjectives. For example, in Standard English, we can say that I am or that he is, but not “I am” or “he is”. This is because the grammar of language requires that the verb and its subject correspond personally. The pronouns I and him are the first or the first. third person, just like verb forms are and are.
The verb must be chosen in such a way as to have the same person as the subject, unlike the fictitious agreement based on meaning.   For example, in American English, the un expression is treated as a singular for the purposes of the agreement, although it is formally plural. . . . .