<?xml version="1.0" standalone="yes"?>
<Paper uid="E91-1055">
  <Title>REFERENCES</Title>
  <Section position="3" start_page="0" end_page="0" type="intro">
    <SectionTitle>
2 GENERATION
</SectionTitle>
    <Paragraph position="0"> Suppose we have the formula given above as a formal paraphrase of (1), and we want to generate an English sentence which corresponds to it. We might hope to use our syntactic/semantic rules &amp;quot;backwards&amp;quot;, looking for something which would generate a sentence and whose semantic component could be made to match the given sequence.</Paragraph>
    <Paragraph position="1"> The final rule we actually used in our analysis of (1) is an elaboration of the standard S ---, NP VP rule which contains a description of how the meanings of the NP and the VP should be combined to obtain the meaning of the NP. Space does not permit inclusion of this rule. The important point for our present purposes is that the representation of the meaning of the S is built up from the discourse representations of the subject and the predicate. The subject and predicate each provide some background constraints, and then their meanings get combined (along with a complex abstraction to the effect that there is some object g which satisfies two properties PO and Pl) to produce a further constraint. The question we want to investigate here is: can we use rules of this kind to generate (1) from the above semantic representation? null The problem is that rules of this kind explain how to combine the meanings of constituents once you have identified them. Given an expression of property theory like the one above, it is very difficult to see how to decompose into parts corresponding to an NP and a VP. So difficult, in fact, that without a great deal of extra guidance it must be regarded as impossible.</Paragraph>
    <Paragraph position="2"> The final semantic representation reflects our beliefs about the best formal paraphrase of the English text, whereas the semantic representations of the components reflect the way we think that this paraphrase might be obtained. Somebody else might decide that they liked our final analysis, but that they preferred some other way of deriving it.</Paragraph>
    <Paragraph position="3"> In view of the number of different ways of obtaining a given expression E as the result of simplifying some complex expression (t E *\[z, P\]), it is simply unreasonable to hope to find the right decomposition of a given semantic representation unless you already know a great deal about the way the linguistic theory being used builds up its representations. Indeed, unless you already have this knowledge it is unlikely that you will even be able to tell whether some semantic representation has a realisation as a natural language text at all.</Paragraph>
    <Paragraph position="4"> If we look again at the knowledge available to our &amp;quot;average NL system&amp;quot;, we see that it will include a vocabulary of lexical items, a set of syntac- 309 tic rules, and a set of semantic interpretations of those rules. It is worth reflecting briefly on the evidence that lies behind particular choices oflexical entry, grammatical rule and semantics interpretation. null The evidence that leads to a particular choice of words to go in the vocabulary is fairly concrete.</Paragraph>
    <Paragraph position="5"> We can, for instance, take a corpus of written English and collect all the contiguous sequences of letters separated by spaces. We can be fairly confident that nearly every such sequence is a word, and that those things that are not words will be fairly easily detected. We would in fact probably want to do a bit better than simply collecting all such letter sequences, since we would want to recognise the connection between eat and eaten, and between die and dying, but at least the objects that we are interested in are available for inspection.</Paragraph>
    <Paragraph position="6"> The evidence that leads to a particular choice of syntactic theory is less directly available. Once we have a vocabulary derived from some corpus, we can start to build up word classes on the basis of looking for words that can be exchanged without turning a meaningful sentence into a meaningless one -- to spot that almost any meaningful sentence containing the word rJalk could be turned into a meaningful sentence containing the word run, for instance. We can then start looking for phrase types and for relations between phrase types. We can perhaps be reasonably confident about our basic classification into word classes, though we may find some surprises, but the evidence for specific phrase types is often in the eye of the beholder, and the evidence for subtler relationships can be remarkably intangible. Nonetheless, there is some concrete evidence, and it has led to some degree of consensus about the basic elements of syntactic theory. You will, for instance, find very few NL systems that do not utilise the notion of an NP, or that do not r~cognise the phenomena of agreement and unbounded dependency.</Paragraph>
    <Paragraph position="7"> The evidence for specific semantic theories, by contrast, is almost entirely circumstantial. We can usually tell whether two sentences mean the same thing; we can usually tell whether a sentence is ambiguous; and we can sometimes tell whether one sentence entails another, or whether one contradicts another. To get from here to a decision that one representation scheme is more appropriate than another, and to a particular translation of some piece of NL into the chosen scheme, requires quite a bit of faith. In order to build a system for translating NL input into some computeramenable representation we have no choice but to make that act of faith. We have to choose a representation scheme, and we have to decide how to translate specific fragments of NL into it and how to combine such translated fragments to build translations of larger fragments. Examples abound. The system that constructed the translation of (1) into the given sequence of propositions in PT is described and defended at length in \[Ramsay 1990\], and we will not recapitulate it here. We note, however, that the rules we use for translating from English into this representation scheme wilt not generate arbitrary such sequences. Only sequences which correspond to the output of the rules we are using applied to the translations we have allocated to the lexical items in our vocabulary will be generated. Tibia is true of all NL s!/stems that translate from a natural language into some formal representation language.</Paragraph>
    <Paragraph position="8"> For any such system, only a fraction of the possible sentences of the representation language will correspond to direct translations of NL sentences, and the only way of telling which they are is to look for the corresponding NL sentence.</Paragraph>
    <Paragraph position="9"> Suppose we wanted to develop a system which used our linguistic knowledge base to generate texts corresponding to the output of some application system. It would be absurd to expect the application program to generate sentences of our chosen representation language, and to try to work from these via our syntactic/semantic rules to an NL realisation. We have no convincing evidence that our representation language is correct; we have no easy way of specifying which sentences of the representation language correspond via our rules to NL sentences; and even if we did have a sentence in the representation language which corresponded to an NL sentence, we would have a great deal of difficulty in breaking it into appropriate components, particularly if this involved replacing a single formula by the instantiation of some abstraction with an appropriate term.</Paragraph>
    <Paragraph position="10"> We suggest instead that the best way to get an NL system to generate text to satisfy the requirements of some application program is for it to offer suggestions about how it is going to build the text, along with explanations of why it is going to build it that way. We therefore supplement our descriptions of linguistic structures with a component describing their functional structure.</Paragraph>
    <Paragraph position="11"> For the rule for S, for instance, we add an element describing what the SUBJECT and PRED are for. We could say that the SUBJECT is the theme and the PRED is the theme, using terms from functional grammar \[Halliday 1985\] for the purpose. A language generation system using the above rule can now ask the application program whether it is prepared to describe a theme and a theme. Admittedly this still presumes that the application program knows enough about the linguistic theory to know about themes and themes, but at least it does not need to know how they are organised into sentences, how they can be realised, or how their semantic representations are combined to form a sentence in the representation language. Furthermore, if the application program is to make full use of the expressive power of NL then it must be able to make sensible choices about such matters, since any hearer will be sensitive to them. If the combination of application program and N L gener- 310., ation system cannot make rational decisions about whether to say, for instance, John ate it or It was eaten by John then they must expect to be misunderstood by native English speakers who are, albeit unconsciously, aware that these two carry different messages.</Paragraph>
    <Paragraph position="12"> Once the application program has agreed to describe a theme and a rheme, the NL system can then elicit these descriptions. Since the rule being used specifies that the theme must be an NP then it can move on to rules and lexieal entries that can be used for constructing NPs and start asking questions about these.</Paragraph>
  </Section>
class="xml-element"></Paper>