Hasty conclusions

The fallacy “Guilt by Association”, discussed in my last blog, is related to two other fallacies, which I have already treated before: argumentum ad hominem or “playing the man” and “jumping to conclusions”. An example of the latter is a fallacy called “hasty generalization” (HG). Because HG is so often found in present-day political discussions, I want to write a few words about it. Hasty Generalization – also called Overgeneralization or Faulty Generalization – is the fallacy that one or several singular cases are seen as exemplary for the group the case belongs to, without further evidence. Or, as Wikipedia tells us: HG is a conclusion drawn about all or many instances of a phenomenon on the basis of one or a few instances of that phenomenon. You could also call it “proof by example”, or better “false proof by example”, for a false proof it is.
When writing this, an example immediately comes to my mind. Recently (case one), near Amsterdam a girl was murdered by a young male asylum seeker when she cycled home at night. Many people, and especially some rightist politicians, reacted with the claim: Asylum seekers are criminals; there is no place for them in the Netherlands. However (case two), a few years ago a similar case happened in the province of Friesland in the north of the Netherlands. Just as in case one, a girl cycling home was found murdered, now not far from an asylum seekers’ centre. Many people reacted that the murderer must have been an asylum seeker, adding that there is no place for asylum seekers in the Netherlands. After some time, the murderer was caught: A local Friesian farmer, living not far from the crime scene. Should we remove then all Friesians (or at least all Friesian farmers) from the Netherlands? Nobody said so in case two (and many of those who uttered the false accusation should have to leave then themselves).
These two cases together are good examples, I think, of why without further information, a single instance cannot be seen as representative for a whole group or population. Just as one criminal Friesian farmer doesn’t “make” all Friesians (or all Friesian farmers) criminals, one young asylum seeker doesn’t “make” all asylum seekers criminals. The generalization is too hasty, namely based on insufficient evidence. In general, we can say that a Hasty Generalization happens
1) when there is a lack of knowledge of the selected example (there was no reason to make a connection between the murder and the asylum seekers’ centre in Friesland);
2) when the selected sample is not representative of the whole group (the murderer in case one is not representative of all asylum seekers, just as the farmer is not representative of all Friesians or Friesian farmers);
3) when both (1) and (2) are true. (see Michael J. Muniz, “Hasty Generalization”)
It is also striking in case one that many people saw the murderer as an asylum seeker, and not as a young man, who happened to be an asylum seeker, or as a man, in the first place. Humans are complicated beings with many characteristics, and why pick out just one characteristic?

In formal terms, the argumentation often follows the patterns:
1)         X is true for A
            There is a connection between X and A.
For instance, the crime scene is near the asylum seekers’ centre, so there is a connection between them.
Or: Asylum seeker A has committed a crime, so asylum seekers are criminal. We need much more information, before we can to draw such a conclusion.
2)         A belongs to population P
            A has characteristic X
            So the whole population P or everybody or many in population P have characteristic X.
For instance, A is a criminal, so everybody in or at least many of his group are criminals.

We see hasty generalizations committed almost every day in politics and the media. In politics, HGs are used to emphasize the extremes of a particular viewpoint. For example, case one (the murder near Amsterdam) is often cited in the present Dutch immigration discussions, however without mentioning case two (the murder in Friesland), which would refute its “conclusion”. We find HG not only in political discussions, but also in other discussions in the media, in advertising to promote particular products, etc. To avoid committing this fallacy, the arguer should take into consideration the amount of justifiable knowledge one might have on a particular subject and whether the selected sample being used in the case is justifiably representative of the group in question. (Muniz) Often this is not simple. Good arguing requires much background knowledge and much insight into the problem at hand. A first step to avoid logical mistakes is to take your time. Avoid hasty steps and jumping to your conclusions.