It has been depicted that pragmatics uses a broader context. It means, in order to do its work properly, a skilful speaker has to count with the effects of the words, sentences and whole utterances that he is planning to deliver. In other words, the speaker presupposes the possible conditions and consequences; and thus, pragmatics works with four tools - reference, inference, presupposition and entailment - which help to decode this effort. For the beginning, it would be helpful to divide these terms into 2 groups - reference with inference and presupposition and entailment - because they are slightly related.
When a speaker says something he uses words which point to entities in the world. He reckons that a listener will recognize what is said because of the clear familiarity of the words and entities in the language they both take control of. The words just refer to concrete entities and thus, the process is called reference. During inference, the speaker is also pointing to the real person or thing; nevertheless, as he/she often does not know the proper or right name he uses expressions such as "that old man" or "this pen on the table" and thus he awaits that the listener will infer which man or pen he has in his mind. It is word pointing that as far as reference and inference are connected, people naturally do this even without knowing they do this (Yule; ch. 3).
The situation with presupposition and entailment is slightly different. Firstly, it is necessary to distinguish between semantic and contextual presupposition. A semantic presupposition is bounded just to the grammatical context, i.e. grammatical structures, e.g. sentence type, in which it occurs; the contextual presupposition, on the other hand, depends also on the context in which the utterance occurs. In other words, semantic presupposition applies naturally to sentences, while contextual one applies to whole utterances (Katz and Langendoen 2). It is just logical that when doing an analysis of the speech not only the semantic presuppositions but mainly the contextual presuppositions must be taken into consideration because only the realm of syntax would not be sufficient and pure lexical and sentence semantics require also the reference to the sociolinguistic context (Keesing 16). This, of course, does not mean that when someone wants to interpret the sentences in context he would abandon semantic presupposition (Katz and Langendoen 15).
So far no mention of entailment has been made. By doing this, it would help also to distinguish it from presupposition. While presupposition is simply the relationship between two propositions which consequently makes a statement, for instance, from the statement "Barrack Obama visited the Czech Republic”, it is clear that there exists a person named Barrack Obama and a state named the Czech Republic. By using presupposition, the awareness of such things comes just from presented facts and neither part of the sentence is considered to be known more than the other. Nevertheless, if someone wishes to stress some kind of information he would do it by using the focusing particular piece of statement either by pitching of the voice on some part of the statement (usually during oral communication) or by it-cleft constructions (more often in written text). With the help of entailment, the previous statement may thus be presented as "it was Barrack Obama who visited the Czech Republic" or "it was the Czech Republic that was visited by Barrack Obama". The former sentence stresses that Barrack Obama visited "something" and presupposes the Czech Republic to be a shared well-known fact, the later, on the contrary, that it was just the Czech Republic that was chosen for the visit by well-known Barrack Obama. It follows that the entailment has more powerful sense than presupposition because the producer of the message decides what should be stressed and what should be taken as a shared knowledge (Yule, ch. 5). And it would be also useful for decoding the political messages.
This thesis does not undertake to offer a thorough description of all kinds of presuppositions and entailments simply due to the fact it does not aim at such descriptions. However, one distinction could be made. According to Grice, it is necessary to differentiate between the so-called conventional and conversational implicature. While the later rests on various principles that govern discourses, the former one consists in the meaning of the words in a sentence and as such it is pure semantic entailment (qtd. in Katz and Langendoen 13). It may be illustrated on the example of the sentence "she was poor but she was honest" where according to the conjunction “but” it is nearly automatically assumed that when a person is poor he must be also unfair and honesty is considered to be something like exception (Leech 10). Nevertheless, the key notion is that this conclusion has been made purely on account of the sentence without knowing any further details about this woman. In a public speech, however, such assumptions should be uttered very carefully and with regard to the customs of particular culture. For this task politicians should rely more on conversational implicature that observe variations of different cultures. As Lakov points out "there are culture-specific assumptions that have to be characterized in order to understanding various aspects of speech acts in a given culture" (qtd. in Keesing 17). If a politician omits this rule he may evoke useless misunderstanding and sometimes even dangerous consequences.
Discourse
Our attention has been so far devoted just to the questions of pragmatics. In the following chapter and sub-chapters, the focus is shifted to the question of discourse and its methods of analysis.
The term discourse is sometimes attributed to any kind of communicated information. This description is not so far from the truth; however, attention should be paid to all of premises which influence real discourse. All of the meanings which accompany the act of communication are joined in order to form clauses, sentences and utterances. Nevertheless, these structures have sense as a well-formed discourse only in particular situational context (Dontcheva-Navrбtilovб; ch. 5). It means, for its analysis, not only syntactic and semantic features are important but also the pragmatic features of particular situation, as it has been stressed in the previous chapter.
Nevertheless, the discourse includes all of its possible forms i.e. spoken or written and also the monologue or the dialogue. The first pair is distinguished under the heading of the medium; the second pair, on the other side, is the result of the nature of the participation during a concrete communicative event and may be bring together under the heading of various aspects of modality. All of these four types have their typical features; nevertheless, it sometimes happens that features that are usually associated with informal dialogic speech are part of a written text, or, on the contrary, when some formal features usually ascribed to writing are incorporated into a public speech (Crystal 69).
Crystal elsewhere tells that "any piece of discourse contains a large number of features which are difficult to relate to specific variables to in the original extra-linguistic context even though the may be felt to have some kind of stylistic value" (63). The analysis thus should be done very carefully in order to catch all contextual features as much as it is possible.
Discourse Analysis
It is not surprising to say that there are various kinds of discourse which may be characterized by various features. However, more intriguing is the question how to recognize these features and how to analyze them. In the following sub-chapter, there are examples of some of the methods of critical discourse analysis.
