(S) = 
(S) = (S) = (S) = | The above propositions in a form of the predicate calculus describe to a certain level the semantics of three kinds of syntactically distinct sentences. The language user is usually able to distinguish each grammatical type by its surface structure. The linguist may now relate these propositions to the syntactic description by means of rules of interpretation. Relating these two views serves as an explanation for the origin of certain grammatical features of the sentence. Using the linguist's syntactic category for the grammatical sentence (S), |
| [+Declarative]S (S) = |
| [+Performative]S (S) = |
| [+Imperative]S (S) = |
| [+Interrogative]S (S) = |
| MultiNet has a relation (NAME) corresponding to my reference functor. The rule for the syntactic feature of declarative thus corresponds to the logical form shown in figure 8. I have also indicated that the abbreviated net for this kind of sentences makes use of suppositional derivation of a situation from an abstract object representing the phrasal of the sentence. MultiNet has only six relations for deriving words and their syntactic structures based on their underlying meaning changes. It manages without such ubiquitous (in English) relations by allowing for numerous adjustments and variations to the signatures of most of the relations it has. We will freely create such ad hoc relations as they are needed. |
Semantics of the syntactic components of a sentence.There are additional rules at a deeper level to describe the interpretation of the other variables. For example, to interpret the syntactic predicate (Pred) in terms of the semantic predicate () of an imperative: |
| [+Imperative]Pred (Pred) = |
| To distinguish the two interrogative sentence types the linguist must be able to designate the different kinds of information functors involved. The yes-no interrogative requires one of the various functors , say, , which is Boolean. The description of a wh-interrogative involves the other class of functors, say, , which are all non-Boolean functors. These different functors of the proposition have values that are objects, numbers (scalar), predicates (sets), or dependent propositions. Each is related to a corresponding syntactic category having the feature [+Interrogative] by means of a rule of interpretation. |
Importance of semantic transformations.These particular rules of interpretation are for sentences of only three syntactic kinds. Such rules for sentences make it tempting to continue to construct some simple one-to-one correspondence between form and meaning. The English speaker, however, finds it easy to come up with myriad counter examples to such simplicity. Often the underlying semantic structure of the author’s sentence does not directly reflect its syntactic form. The uses in the table in §3 are not clearly defined; each one flows into the next. Nor does the same use always correspond to the same syntactic structure. To state these myriad relationships the linguist will need to give a like number of semantic transformations. |
Rhetorical questions.One of the more common transformations is the so called rhetorical question. This sentence has the syntactic form of a question but the import of an emphatic assertion. The linguist may describe this phenomenon by means of a sentence level semantic transformation. In a certain situation, when the language user knows that the speaker does not expect an answer, he interprets the wh-interrogative as if it were a corresponding emphatic specific declarative. |
| (1) | a. | Who would do such a thing? | |
| b. | How long must I wait? |
| The sentences in (1) are questions, albeit rhetorical questions. Their interpretation is as if they had the form of the declarative that contained the full answer. |
| (2) | a. | Only a certain person would do such a thing! | |
| b. | I must wait so very long! |
| Now for purposes of illustration suppose that there was a more atomic description of such sentences as in (1) that included sufficient features, e.g., [+Emphatic,+Specific]S to allow the language user to infer its intended interpretation. These features would make it possible for the linguist to express the relationship between the two different syntactic types formally. Here is roughly how a rhetorical question transformation might look: |
| RHETORICAL QUESTION (RQ): | |
|---|---|
| base: | [+Interrogative,+Rhetorical]S . . . [+Interrogative]Cat |
| transform: | [+Declarative,+Emphatic]S . . . [+Specific]Cat |
| The condition for this interpretation, the feature [+Rhetorical], arises from a condition on the interrogative speech act — what linguists call a felicity condition. The means of describing the syntactic sentence type in the previous section was to give various felicity conditions. To ask a question felicitously the person asking must not know the answer beforehand: | |||
| This means that if the hearer can conclude from the situation that the author does in fact know f(x), then the feature [+Rhetorical] might appear on the sentence. The feature arises from the semantic functor asserting the truth of the proposition | |||
| which holds in the larger context of the utterance. | |||
| [+Interrogative,+Rhetorical]S r |
| These features trigger the RQ transformation as formulated above. The transform yields the new interpretation of the sentence as emphatic with the syntactic category corresponding to the interrogative being interpreted as specific. |
Polite Requests.Linguists can appeal again to a sentence level semantic transformation, when they try to account by rule for the various circumlocutions of an imperative used to make a (polite) request. |
| (1) | a. | Will you please come here? | |
| b. | Would you please come here? |
| The feature of second person on the subject noun normally grammaticalizes a hearer as agent (cf. §11-1-3). Future tense normally grammaticalizes future time (cf. §17-5-6). The language user can normally infer both these facts from the form of an imperative sentence. The close similarity of meanings between the sentences in (1), however, suggests that the grammatical tense here is not conveying the aspect of time. The inquiry appears to be the same in both versions. The speaker is asking about the willingness of the hearer to perform the action. The answer would specify one of the felicity conditions on any request. |
| This is the reason for the second person interrogative. This form asks about the desire motivating its interpretation as a request — polite imperative. The language user may interpret yes-no interrogatives (cf. sentences like those in the fifth paragraph of §6-3) which ask about the desire of a second person subject, as a corresponding imperative. |
| POLITE IMPERATIVE (POL-IMP): | |
|---|---|
| base: | [+Interrogative,nPolite]S . . . [2Person]NP |
| transform: | [+Imperative,nPolite]S . . . [+Request]Pred |
| In this case the linguist may derive the feature of [1Polite] from the first felicity condition on a request and the fact that the interrogative asks about the desire of the hearer. Standard rules already given would suggest the following normally applicable rule of interpretation: |
| [+Interrogative,1Polite]S…[+2Person]NP r |
| In this case, however, the hearer would not use the normal rule of interpretation. Usually with an imperative the pro-form sentence for the predicate response would be OK (as a speech act) or the actual performance of the action described, and that would seem to be the ready interpretation in force here. |
| [+Imperative,1Polite]S…[+Request]Pred r |
Additional levels of politeness.Other degrees or types of polite requests take advantage of additional felicity conditions. |
| (1) | Do you want to come here? |
| …… |
| (2) | Can you come here? |
| …… |
| (3) I would like you to come here. |
| These conditions suggest adjustments to the interpretation rules for requests with different kinds or degrees of politeness. Except for (3) they all fit the general pattern of POL-IMP. The linguist would have to devise something like the following for the most polite form: |
| VERY POLITE IMPERATIVE (0-POL-IMP): | |
|---|---|
| base: | [+Declarative,0Polite]S . . . [1Person]NP |
| transform: | [+Imperative,0Polite]S . . . [+Request]Pred |
More semantic transformations.As the linguist proceeds with the analysis and description by means of semantic transformations, one finds the need becoming more accute to refer to features that must originate on subordinate components of the base. It may also be desirable to define the transform more strictly by referring to structures of substructures. The reason for simplifying the statements of the transformations in these sections has been to illustrate the fluctuating nature of the relationship of form to meaning. |
