Fallacies and Argument Appraisal by Tindale | Excerpt

The following excerpt about the post hoc fallacy (a specific type of correlation-causation fallacy) is from Professor Christopher W. Tindale's Fallacies and Argument Appraisal.

The post hoc fallacy involves attributing a causal relationship where none exists. Other problems related to causal reasoning involve mistakes about an actual causal link. Two such errors involve types of misidentification of causes. In the first instance, we may falsely identify X as the cause of Y when on closer inspection a third factor, Z, is the cause of both X and Y. In the second case, we may confuse a cause and effect...

Case 9C
The decision of large numbers of women in the last decades of the twentieth century to abandon the home and instead pursue careers in the workplace has led to a decrease in the birthrate, since fewer are making the decision to stay home and be mothers. 

There is something a little circular about the reasoning here, but our interest is in whether this expresses a plausible causal relationship. Presumably, we could find correlations in studies that show that birthrates decreased during the same period that more women chose careers over homes. But if we look behind these two factors, we see that things like advances in medical technology, specifically in contraception, have made possible both effects. It is not so much that choosing a career over motherhood has caused a decline in the birthrate, but that widely available access to contraception has allowed women both to enter the work force in greater numbers and to plan their families better. In all cases in which causal relationships are suggested, we should look beyond them to the wider context in which they arise to see whether there could plausibly be a common cause that can account for both factors...

The second type of misidentifying a cause may be a little easier to detect if we are alert to its possible presence. To argue, for example, that the presence of a women’s shelter has led to an increase in crimes against women because the statistics show more reported assaults after the shelter is established than before (that is, the correlation is there) misses the causal relationship. It is the sexual assaults that create the need for, and thereby cause, the existence of the shelter, and once women have a safe venue to report to, there is an increase of reported assaults. Thus, it is not that X (the women’s shelter) causes Y (the increase in assaults), but that Y (the assaults) causes X (the shelter).


Fallacies and Argument Appraisal by Christopher W. Tindale
Copyright © 2007 Christopher W. Tindale.  Reprinted with the permission of Cambridge University Press.