Besides genuine syllogisms, there may be syllogisms that look genuine but are not; and since an enthymeme is merely a syllogism of a particular kind, it follows that, besides genuine enthymemes, there may be those that look genuine but are not.
1. Among the lines of argument that form the Spurious Enthymeme the first is that which arises from the particular words employed.
(a) One variety of this is when-as in dialectic, without having gone through any reasoning process, we make a final statement as if it were the conclusion of such a process, 'Therefore so-and-so is not true', 'Therefore also so-and-so must be true'-so too in rhetoric a compact and antithetical utterance passes for an enthymeme, such language being the proper province of enthymeme, so that it is seemingly the form of wording here that causes the illusion mentioned. In order to produce the effect of genuine reasoning by our form of wording it is useful to summarize the results of a number of previous reasonings: as 'some he saved-others he avenged-the Greeks he freed'. Each of these statements has been previously proved from other facts; but the mere collocation of them gives the impression of establishing some fresh conclusion.
(b) Another variety is based on the use of similar words for different things; e.g. the argument that the mouse must be a noble creature, since it gives its name to the most august of all religious rites-for such the Mysteries are. Or one may introduce, into a eulogy of the dog, the dog-star; or Pan, because Pindar said:
O thou blessed one!
Thou whom they of Olympus call The hound of manifold shape That follows the Mother of Heaven: or we may argue that, because there is much disgrace in there not being a dog about, there is honour in being a dog. Or that Hermes is readier than any other god to go shares, since we never say 'shares all round' except of him. Or that speech is a very excellent thing, since good men are not said to be worth money but to be worthy of esteem-the phrase 'worthy of esteem' also having the meaning of 'worth speech'.
2. Another line is to assert of the whole what is true of the parts, or of the parts what is true of the whole. A whole and its parts are supposed to be identical, though often they are not. You have therefore to adopt whichever of these two lines better suits your purpose. That is how Euthydemus argues: e.g. that any one knows that there is a trireme in the Peiraeus, since he knows the separate details that make up this statement. There is also the argument that one who knows the letters knows the whole word, since the word is the same thing as the letters which compose it; or that, if a double portion of a certain thing is harmful to health, then a single portion must not be called wholesome, since it is absurd that two good things should make one bad thing. Put thus, the enthymeme is refutative; put as follows; demonstrative: 'For one good thing cannot be made up of two bad things.' The whole line of argument is fallacious. Again, there is Polycrates' saying that Thrasybulus put down thirty tyrants, where the speaker adds them up one by one. Or the argument in the Orestes of Theodectes, where the argument is from part to whole:
'Tis right that she who slays her lord should die.
'It is right, too, that the son should avenge his father. Very good: these two things are what Orestes has done.' Still, perhaps the two things, once they are put together, do not form a right act. The fallacy might also be said to be due to omission, since the speaker fails to say by whose hand a husband-slayer should die.
3. Another line is the use of indignant language, whether to support your own case or to overthrow your opponent's. We do this when we paint a highly-coloured picture of the situation without having proved the facts of it: if the defendant does so, he produces an impression of his innocence; and if the prosecutor goes into a passion, he produces an impression of the defendant's guilt. Here there is no genuine enthymeme: the hearer infers guilt or innocence, but no proof is given, and the inference is fallacious accordingly.