There is a rule about arguments which tells us that if a term in the conclusion refers to the whole of its class, then the evidence pointing to that conclusion must also have told us about the whole class. We cannot reach a conclusion about ‘all estate agents’, for example, unless we start with some knowledge which applies to all of them. To know that some estate agents are guilty of this or that practice will not justify us reaching conclusions about all of them. Arguments which break this rule are said to commit the fallacy of illicit process.
All tax-collectors are civil servants, and all tax-collectors are bullies, so all civil servants are bullies.
(Too harsh. There may be some somewhere who are just a little overbearing. The fallacy is that we refer to all civil servants in the conclusion, but the premise only tells us that tax-collectors are some of them.)
The argument which uses illicit process has to be fallacious because it makes unsupported claims. Although the premises talk only about some of a class, the conclusion introduces for the first time the rest of that class. In other words, we try to reach conclusions about things we have no evidence on, and commit a fallacy by doing so.
There is another version of illicit process which is harder to spot:
All cyclists are economical people, and no farmers are cyclists, so no farmers are economical people.
(This appears to fit the observed facts, but there is a fallacy. We could just as easily have said ‘All cyclists are mortals’. This would give the distinct impression that big fat farmers would be driving their big fat cars for ever.)
The source of the fallacy in this example is that the premise tells us that cyclists are some of the class of economical people. The conclusion, on the other hand, tells us that the entire class has not a single farmer in it. Again, the fallacy is illicit process.
These terms which cover the whole of their class are called ‘distributed terms’, and there is a rule for finding them. Universal, which talk about ‘all’ or ‘none’, have distributed subjects; negatives, which tell us what is not the case, have distributed predicates. In the example above, the term ‘economical people’ is distributed in the conclusion, since it is the predicate of a negative statement. In the premise, however, it is undistributed, being neither the subject of a universal nor the predicate of a negative. It sounds complicated, but the rule makes it simple. You will soon be seeing which conclusions try to cover all of a class without any information to justify it. To dazzle your friends totally, you should call the fallacy illicit minor when the subject of the conclusion is unjustifiably distributed, and illicit major when the predicate of the conclusion is so treated.
To use illicit process requires a good deal of homework. You should deploy it in support of conclusions which look plausible but have the minor technical drawback that you cannot prove them. Your expertise at illicit process will enable you to construct arguments based on what some of the class do, and slide smoothly into conclusions about all of them.
Some Australians are pleasant fellows, and some con-men are n pleasant fellows, so some Australians are not conmen.
(Who knows? It may even be true; but it takes a lot more than this to prove it.)