Notes and questions 1 The syllogism is reliable only in certain situations. Because natural language is a system in itself, there are a number of dangers when it is used to give expression to other knowledge systems: (a) a mouse eats cheese; (b) a mouse is a word of one syllable; (c) therefore, words of one syllable eat cheese. In particular, the words employed must keep their same signification in the major and minor premises. If they do not, then the conclusion can be absurd, as the example indicates. Also care must be taken not to confuse genus and species: (a) cats eat meat; (b) cats are animals; (c) therefore, animals eat meat. This conclusion might not be quite as absurd as words eating cheese, but it is equally wrong, in as much as it does not accord with reality. Not all animals eat meat. In addition, there is the problem of the missing premise (tertium non datur): (a) smoking causes lung cancer; (b) Socrates died of lung cancer; (c) therefore, Socrates was a smoker. The missing premise is that lung cancer has other causes as well as smoking. 2 Harbour Assurance shows how the syllogism can prove particularly weak in law in as much as the structure or model is subject to little external restraint and thus proves a ‘movable goal post’. Intellectual systems can, in short, contain uncertainties and ambiguities. Does Harbour Assurance indicate that logic is simply a form of argumentation? Or does it indicate that legal reasoning should never be seen as a matter of logic as such, but a matter of what premise is to be adopted?
Pages 1

Robert Goff LJ:… We are concerned in the present case with the construction of certain words, viz, ‘without having the consent of the owner’, in their context in a particular sub-section of a criminal statute. However, the concept of consent is relevant in many branches of the law, including not only certain crimes but also the law of contract and the law of property. There is, we believe, danger in assuming that the law adopts a uniform definition of the word ‘consent’ in all its branches…