Types of Causality

By Bradley Wright

(Part II of a series)

In my last post, I wrote of basic ideas about causality. Sociologists most readily assume that one thing in the social world causes another when the cause correlates with the effect, occurs before it, and there’s a plausible, non-spurious causal mechanism. 

In this post, I develop the idea of causality more by talking about its different types. Yes, just as there are different types of cars, ice cream, and television reality shows, there are different types of causality. Knowing these distinctions in causality allow us to recognize causality more readily and to think about it more deeply… and you thought you’d have nothing to do during your summer vacation!

1) Nomothetic vs. Idiographic . The first distinction involves two words no one has ever heard of: nomothetic and idiographic (they come from the Latin phrase “really confusing”). They regard how many cases are being explained—many or just one. 

Nomothetic means a causal relationship is assumed to happen among many cases. (Sociologists usually study people as cases, but these principles apply to non-people cases like bears or stars or battles). With nomothetic causation, some cause has some effect on lots of people. This is what sociologists do almost all the time, and it’s so routine, we can forget that there is anything else.

Idiographic causation, however, involves just one person (or case). Basically, you’re saying that a cause has an effect for one person or thing, and you don’t know, or don’t care, whether it affects others. 

Now, why would anyone want to go idiographic? Well… historians might want to explain why a specific event happened. For example, why did the American Civil War occur? Different historians might have different explanations, but they are not trying to explain all wars (this would be nomothetic) or even all civil wars (again, nomothetic), but rather why this particular war happened.

2) Deterministic vs. Probabilistic . The next distinction of causality is fortunately easier to pronounce, but it still identifies a type of causality that people sometimes miss. This distinction regards whether a cause happens every single time or just some of the time.

Deterministic causation occurs when every time you have a cause, you have an effect. For example, every time you cool pure water to 32 degrees, it freezes into ice. (Okay, I suppose any science types reading this would find an exception, e.g., maybe water under high levels of pressure doesn’t freeze or maybe it acts differently in outer space, but work with me here).

Sociologists rarely, if ever deal with deterministic causation. Instead, we’re all about probabilistic causation. This happens when a cause sometimes brings about an effect, but not always. A classic example is smoking and lung cancer. Not everyone who smokes gets lung cancer, but it definitely increases the likelihood of it, so we call smoking a cause of lung cancer.

Sometimes people confuse this distinction, and they proclaim that if a cause is not deterministic, it is not really a cause. For example, someone might argue that since their grandfather smoked a pack a day and died at natural causes at 100 years old smoking doesn’t cause lung cancer.

In this sense, probabilistic causation is more difficult to deal with (and I think more interesting) because it does have lots of exceptions, but, it’s still causation. 

3) Necessary vs. sufficient. This next distinction is rather tricky. It’s the one that that most sociology students have to stop and really think about when they’re answering a midterm question. It regards whether a particular cause is needed to bring about an effect and if that cause is enough by itself.

Necessary causation occurs when you absolutely, positively must have a certain cause to observe a corresponding outcome. Simply having that cause may not be enough to have the effect, but you definitely need that specific cause to have any hope 

of observing the effect. A simple example: You need air to be alive, so air is a necessary cause for life. Air, by itself, might not be enough, for other things can come into play. Maybe you were just bitten by some crazed monkey that has Ebola, and you’re going to be dead by sundown no matter how much air you breathe.

Sufficient causation means that a particular cause is enough to have an effect, though other causes could bring that effect about as well. Eating a pint of Ben & Jerry’s ice cream every day is enough to make you fat just by itself (assuming that the rest of your diet stays the same). It is a sufficient cause of weight gain. (By the way, I just tried their flavor “Visual Whirled Peace” today—outstanding). But, there are also other possible causes of weight gain, such as Haagen Dazs and Bryers ice creams.

Just to keep you on the edge of your computer chair, let me tell you that I plan to write about several other distinctions in causality in my next post. How sweet it is…

Comments

[Carla]

Brad, this series of posts is really interesting and useful! Thanks. Except, now I really want ice cream and I shouldn't eat it, so, no thanks.

[simone rozzi]

Hi I like this post and the distinctions you made about causality. Inoticed that you used the word causality and causation interchangeably, do you think they are sinonimous?

[C W]

This s really helpful info!

[Lauren]

Thanks for this interesting information! I never even thought that it was possible to go so deep into cause and effect and this is really great to think about!

[Lung Cancer Info Guide]

I do disagree with you. Smoking and lung cancer are associated with each other.Now 80% of lung cancers are due to smoking.

[neg]

this info helped me a lot..thanks!

[liz]

I have an exam for my doc research class today and this really helped! Thank you so much for posting!