The post hoc ergo propter hoc fallacy assumes a causal relationship between two events or states based on chronological succession. It is a type of non sequitur.
According to such reasoning, if event B follows event A, then A caused B. However, of course, chronology doesn’t imply causation. If something comes after something else, it doesn’t necessarily mean the first thing caused the second thing.
Post hoc ergo propter hoc translates from Latin as “after this, therefore because of this.” The fallacy assumes a causal relationship between two events or states based on chronological succession. According to such reasoning, if event B follows event A, then A caused B.
We find the concept for such a fallacy in Aristotle’s Rhetoric:
“Another fallacy consists of taking what is not the cause for the cause, as when a thing has happened at the same time as, or after, another; for it is believed that what happens after is produced by the other, especially by politicians. Thus, Demades declared that the policy of Demosthenes was the cause of all the evils that happened, since it was followed by the war” (Aristotle, Rhetoric, Book II, Chapter 24, Section 8).
Such arguments are considered fallacies because we know that temporal succession doesn’t necessarily imply causation.
Just because B followed A, it doesn’t mean that A caused it. Such arguments fail to take into account other causally relevant factors and focus on one factor.
Some philosophers distinguish between different types of fallacies. Thus Pinto (2001, p. 59) identifies three types:
- causal interpretation of particular events (concluding that particular event A caused a particular event B from the fact that B followed A)
- inference from correlation to cause (concluding that there is a causal relation between two different types of events because of their temporal succession)
- Causal generalization from one sequence of events (concluding that there is a causal relation between two event types A and B from the fact that a single instance of A was followed by an instance of B)
It is, however, important to note that if event B always occurs after event A, we are close to establishing a causal law.
The precise difference between necessary connections and fallacious assumptions has been a subject of debate among philosophers, at least since David Hume (Hamblin, 1970/2022).
For example, I see the window shatter when the ball hits it. I’m right not to consider alternative causes and hypotheses. I will immediately assume a causal relationship. Indeed, temporal succession doesn’t prove causation but assuming a causal relationship is often reasonable.
- Wrong economic assumptions: Arguing that unemployment decreased in America because Switzerland eliminated its gasoline tax a year before.
- Marriage counselling and divorce: Assuming that going to a marriage therapist causes divorce because a lot of people who go to couples therapy get divorced shortly after. The reality is that those relationships were already on the rocks when they went to the therapist, hence the correlation.
- Sporting superstition: A tennis player won two games in a row, and also happened to eat chicken the night before each game. From then on, he decides to always eat chicken before his games because it seems to cause him to win.
- Seeing a black cat: An old superstition states that if you see a black cat, you will have a bad day. Subsequently, a man sees a black cat and has a bad day. He remembers the old superstition and decides that it was the black cat that caused his bad day today.
- Assuming false patterns in the stock market: Stock market advisors make predictions based on a few coincidental indicators which, in the past, just happened to precede a rise or fall in prices, despite there’s no logical reason for there to be causation between the two events.
- Blaming a political party for the weather: Arguing that a drought was caused by the ruling political party because that party was in power when it happened.
- New laws and future unrelated events: Arguing that, as in Aristotle’s example, some policy is the cause of all evils that happened right after it was passed, only because of temporal succession.
- Astrology: Predicting what will happen to a person this week based on their zodiac sign and the movement of stars through the sky last week.
- Ice-cream leads to crime: Assuming that eating ice cream causes a rise in crime rates because a rise in ice cream consumption is often followed by a rise in crime. This phenomenon can alternatively be explained by a common cause: the summer.
- The cleaner stole my wedding ring: A man last remembers seeing his wedding ring on the bedside table in his hotel room. When he gets back from lunch, he notices that the wedding ring is not there, and also that the cleaners have been through the room. Logically, he assumes that the cleaner stole his ring. While this is a logical assumption and worthy of further investigation, he may need to find proof to avoid falling into the post hoc ergo propter hoc fallacy and making a fool of himself – especially because a week later his wife remembers that she pocketed it for safekeeping!
Hansen (2020) gives the following example of post hoc ergo propter hoc reasoning:
“Unemployment decreased in the fourth quarter because the government eliminated the gasoline tax in the second quarter.”
Such reasoning is potentially fallacious because the decrease in unemployment that took place after the elimination of the gasoline tax may have been due to other causes like an increase in demand for products in the fourth quarter.
Another possibility is that these two events occur in succession because they both have a common cause.
Keep in mind that the statement above may very well be true. It’s not a formally invalid argument.
This argument is fallacious not because it is formally invalid, but because it makes unwarranted assumptions.
The whole field of astrology is partly based on such informally fallacious arguments as post hoc ergo propter hoc.
Some event B followed event A (the exact date when a person was born). It is, therefore, assumed that the people born on the same exact date will also experience event B in their lives.
A is taken to be the cause of B when no real reason for that assumption exists.
This has led to a range of predictions like “I’m having trouble communicating with my husband because mercury is in retrograde.” Here, the assumption is that the position of mercury in the sky caused communication issues.
A famous example of post hoc ergo propter hoc reasoning comes from David Hacket Fischer and is quoted in Woods and Walton’s (1977) paper on arguments:
“On the fatal night of the Doria’s collision with the Swedish ship Grisholm, off Nantucket, the lady retired to her cabin and flicked the light switch. Suddenly there was a great crash and grinding metal, and passengers and crew ran screaming through the passageways. The lady burst from her cabin and explained to the first person in sight that she must have set the ship’s emergency brake” (Fischer, 1970, p. 166).
The passenger concludes that the switch she flicked set off an emergency brake.
She assumed that something that happens on trains might happen on a ship. She also assumed that a light switch was the emergency brake because the ship suddenly stopped after she flicked it. Ships don’t have brakes.
Even if they did, it’s not likely that the emergency brake would be located in individual cabins and shaped like light switches.
This is an example of fallacious reasoning for many reasons, but certain factors, like a sign above the light switch saying “Use only in the event of an emergency,” would make her argument sound much more reasonable (Pinto, 2001, p. 61).
People sometimes assume that going to a marriage therapist causes couples to get divorced.
Let’s assume that more couples who go to a marriage therapist subsequently get divorced than the average population (event B tends to follow event A).
So, people assume a causal relationship. An alternative explanation is that people who already have a rough marriage go to couples therapy and people who need such help are more prone to eventually getting divorced.
Fallacies are deceptively bad arguments (Hansen, 2020). Aristotle and the early nineteenth-century logicians John Stuart Mill and Richard Whately warned us about the mistakes we might make when constructing arguments.
The most common such mistakes are fallacies. Since the 1970s, the importance of understanding how fallacies work has been widely acknowledged (Johnson & Blair, 1993).
For modern logicians, it is common to distinguish two categories of fallacies: formal and informal.
- Formal fallacies are those readily seen as based on invalid logical forms. Examples of formal fallacies include the ad hominem argument, strawman fallacy, and appeal to ignorance.
- Informal fallacies are also often invalid, but their weaknesses are not formal. Examples of informal fallacies include the genetic fallacy and the fallacy of composition.
Such fallacies may sometimes be reasonable and sometimes not. Post hoc ergo propter hoc is one example of an informal fallacy associated with causation.
Post hoc ergo propter hoc is a form of argument which assumes a causal relationship between two states or events based on chronological succession. Such inferences may sometimes be reasonable to make and sometimes not. It is, however, necessary to understand how they work so that we don’t commit fallacies when constructing arguments.
Aristotle. (1926). Vol. 22. In Aristotle in 23 Volumes. J. H. Freese Trans. Harvard University Press. http://www.perseus.tufts.edu/hopper/text?doc=Perseus:text:1999.01.0060
Hamblin, C. L. (2022). Fallacies. Advanced Reasoning Forum. (Original work published 1970)
Hansen, H. (2020). Fallacies. In E. N. Zalta (Ed.), The Stanford Encyclopedia of Philosophy (Summer 2020). Metaphysics Research Lab, Stanford University. https://plato.stanford.edu/archives/sum2020/entries/fallacies/
Johnson, R. & Blair, J. A. (1993). Logical Self-Defence. 3rd Edition. McGraw-Hill Ryerson.
Pinto, R. C. (2001). Post Hoc, Ergo Propter Hoc. In R. C. Pinto (Ed.), Argument, Inference and Dialectic: Collected Papers on Informal Logic with an Introduction by Hans V. Hansen (pp. 56–63). Springer Netherlands. https://doi.org/10.1007/978-94-017-0783-1_6
Woods, J., & Walton, D. (1977). Towards a Theory of Argument. Metaphilosophy, 8(4), 298–315. https://doi.org/10.1111/j.1467-9973.1977.tb00283.x
Dr. Chris Drew is the founder of the Helpful Professor. He holds a PhD in education and has published over 20 articles in scholarly journals. He is the former editor of the Journal of Learning Development in Higher Education.