The Art of Rhetoric, Aristotle
"It is also clear that its function is not persuasion. It is rather the detection of the persuasive aspects of each matter and this is in line with all other skills."
"The genres of rhetoric are three in number, which is the number of the types of audience. For a speech is composed of three factors -- the speaker, the subject and the listener -- and it is to the last of these that its purpose is related. Now the listener must be either a spectator or a judge, and, if a judge, one either of the past of of the future. The judge, then, about the future is the assembly member, the judge about the past is the juror, and the assessor of capacity is the spectator, so that there must needs be three types of rhetorical speech: deliberative, forensic, display."
"It is clear from what has been said that it is in the first place necessary to be in possession of premisses in regard to these matters. (Evidence, probability and signs are the premisses in rhetoric.) Deduction in general is derived from premisses and the enthymeme is a deduction consisting of the premisses just mentioned. Moreover, impossibilities can neither be about to be done nor have been done -- only possibilities can; neither can things that have not happened have been done nor can those that will not happen to be destined to be done. So the deliberative, forensic and display orators must all alike possess premisses of the possible and impossible and of things having happened or not and being about to happen or not. Furthermore, as all speakers, both praising and denouncing, both urgning and deterring, both in prosecution and in defence, seek to demonstrate not only the points that we have mentioned but also that the advantage or disadvantage, the glory or the shame, the justice or the injury are either great or small, either describing them in themselves or comparing one with another, it is clearly also necessary to be in possession of premisses connected with absolute and relative size and smallness, both in general. and in particular, such as which is the greater or lesser advantage or just or unjust deed, and so too with the other aspects."
"Praise and deliberations have a common form. For the things that one might put forward in giving advice would become encomia by a change in the expression. Since, then, we understand what we should do and of what sort we should be, we must take these things said as suggestions and change and adapt their expression; for instance, that we should pride ourselves on the effects not of chance but of ourselves is the way to make the point as a suggestion and the way to make it as praise is as follows: "He is proud not of the effects of fortune but of himself.' So, whenever you wish to praise, see what suggestion you would make, and whenever you wish to make a suggestion, consider how you would praise."
"There are three causes of the speakers' themselves being persuasive; for that is the number of the sources of proof other than demonstration. They are common sense, virtue, and goodwill."
"Those who share one's pleasure in good things and one's pain in painful ones for no other reason than for the sake of their friend. For all men are happy when what they want occurs, and are sad when the reverse happens, so that pains and pleasures are an indication of wishing."
"And anger is curable by time, but hatred not. The former is a pursuit of pain, the latter of evil; for the angry man wants to witness the punishment, but for the hater it makes no difference, since painful things are all perceptible, while those which are particularly bad (injustice and stupidity) are least perceptible. For the presence of wickedness causes no pain. And anger is attended by pain, the other not; for the angry man is in pain, but not the hater. And and angry man might feel pity in many circumstances, but the other in none; for the one wants the man with whom he is angry to suffer in return, while the hater wants his enemy not to exist."
"Maxims give great assistance to speeches, for one thing, through the stupidity of the listeners; for they are delighted if someone in generalizing should arrive at opinions that they hold in the particular case. The following will clarify what I am saying, and also how one should hunt for maxims. The maxim, as has been said, is a general assertion, and the listeners are delighted when a point is generalized which they happen to presuppose in the particular case. For instance, if someone should happen to have bad neighbours or children, he would accept someone's saying that there is nothing worse than neighbourhood or that there is nothing more foolish than child-raising. So one should guess at the sort of opinions that the audience happen already to have presupposed, and then speak in general about them."
"Those by induction from the similar, either one of several, whenever one takes the general point and deduces the particulars, ar that by example. Those by necessity and what is always the case are those by proof, and those from the general or the particular, either real or not, are those by indication; and since probability is not what always happens but what happens for the most part, it is obvious that this sort of enthymeme can be refuted by bringing an objection, but the refutation is sometimes apparent and not always true; for the objector does not refute by showing that it is not probable but that it is not necessary. For this reason, one can always gain a greater advantage in defence than in prosecution through this trick. For the prosecutor demonstrates by probabilities, and it is not the same to refute something as being improbable and as not being necessary (and what is for the most part always admits of objection), but the judge thinks that if the argument is refuted in this way then it is either not probably or not for him to judge, and he is tricked, as we said (for he should not judge only from necessities but also from probabilities, since this is to judge by the best opinion), and it is therefore not sufficient to refute an argument as not being necessary but one must also refute it as not being probably."