Monday, January 9, 2012

Causes vs. Background Conditions

Consider a typical case of causation. Lucy strikes a match to light her grill. It's a beautiful sunny day and there's not a hint of wind. Not surprisingly, the match ignites and she is able to start her grill. I take it that it is the striking of the match that causes it to light.

But there are (what some might call) background conditions that are required for the match to light, like the presence of oxygen and a lack of rain and gusty winds. If the match were wet, then it wouldn't light. And if it were too windy, it wouldn't light. So here's an interesting question: Do the background conditions (also) cause the match to light? This might sound odd. We don't say things like "The striking of the match and the presence of oxygen caused the match to light."

Various theories of causation seem to entail that we should say that these background conditions count as causes (See, for example, David Lewis's view in his essay "Causation" and J.L. Mackie's view in his book The Cement of the Universe, among others). Some folks take the above to be a direct challenge to such theories while others bite the proverbial bullet and simply insist that such background conditions are indeed causes. What do you think about this?

If you think the background conditions are not causes, how exactly do you distinguish between the two and explain why these background conditions aren't causes?

If you think they are causes, then how do you explain our practice of not treating them as such? Why are so many people tempted to say that there's a metaphysical difference between causes and mere background conditions? How do you show that this is a mistake?

16 comments:

Brianna Price said...

I do not think that the back ground conditions are causes for the match to light. I think that they are conditions that could prevent a match not to light just like there are conditions to make the match light. If there is wind then that could be prevented with a cover of your hand so if the wind is present that does not mean the match cannot light because their are ways to block out the wind. I believe a person should take such conditions into effect and modify how that person is to go about lighting that match. Back ground conditions have nothing to do with the object itself even though the background condition could have some effects as to how the match responds but there is the fact that the person can still modify how to go about lighting a match in that condition.

Anonymous said...

If the match was constructed to not be used in windy places or underwater (anaerobic or non-oxygenated environment) then no. It could be said that the ideal environment for lighting a match in fact PRODUCED the match itself as an answer to the said background conditions and the need for fire. Because the match is man made, it was envisioned and implemented for those conditions. Therefore the conditions caused the match, but the girl caused the match to light. It requires the girl to light the match because in unfavorable conditions, the girl could never light the match to begin with.

It was the need for fire, which caused the match to light. If people lived underwater and still needed fire, there certainly would not be any matches, but some other form of starting fire that fits with the context.

Kelly D said...

I think that background conditions are certainly causes. Classifying events as background conditions or causes is the tricky part. At first I am usually tempted to make the distinction in relation to time. So the event that is considered the cause is the one that happens closest in time to the effect. This categorizes all other events as background conditions. This is faulty because of simultaneous causation. Also, perhaps the actually causal event doesn't happen closest in time to the effect? Maybe there are background conditions intermingling before and after the main causal event. But what is a causal v. a non-causal event? This topic always leaves me with more questions than answers! I'm also tempted to make the distinction between background conditions and a causal event by determining which event was absolutely necessary for the effect to occur. The conditions that could vary would be those that are background conditions while the conditions that are vital for the effect would be considered the cause.

Sherry Troutman said...

I think they are causes. Where there's a cause there's also an effect. Whether it be that the match lights or does not is the effect. The causes would be the conditions causing the effect. However, you could always modify the causes to work in your favor of lighting the match. Like holding the match under an umbrella to keep the rain away from the match if it is rainy, and cupping your hand around the match to keep the wind away if it is windy, no matter what the cause--an effect must always be present.

There is no difference between the back ground conditon and the cause--some people may disagree and say that there are but the conditions cause the effect to occur and the causes are the conditions causing the effect to occur so on a technicality if you describe it like that--there are no differences. Just because the cause and background conditions are the same--the reason for them not being treated as such is because conditions or causes can always be modified to get the desired effects.

Jacob Shepherd said...

I believe that background conditions are just causes that are constantly in action. The cause of my writing this blog comment is that I am alive because my brain is functioning (mostly). My brain is functioning because it is receiving oxygen via my bloodstream (if I'm not mistaken, that is...). Background conditions such as oxygen, lack of wind, and the energy used for striking the match are definitely causes of the match lighting. They caused the woman to exist in the first place, since her parents could not have existed without the presence of oxygen, etc. So, these background conditions are not expressed by volition (as the lighting of the match is), but they are nevertheless active in the lighting. Humans just take for granted that elements such as dryness and air are causes for doing some of the things we do. Or, rather, that oxygen, etc. are the causes for our causes. Air and lack of wind cause us, in a lot of ways, to do things like strike matches. They cause us to do things like that because they cause feelings such as the desire to grill or light a fire, which are caused by maybe hunger or a desire to have some fun and what not. Background conditions, then, are causes of causes, unless I'm mistaken of course.

S. Vetterly said...

I think that the background conditions are causes. However, they are not considered as such by the average person because all that is observed is the direct "cause and effect" brought about by the striking of the match and the ignition of the match. The average person does not take the time to realize all the conditions that must be true and present (such as the presence of oxygen, the lack of water on the match, correct properties of whatever is being used to strike the match, etc).

Another example of such a problem is hitting someone with a snowball. While the main cause and effect of the action may simply be the movement of the arm, other things come into play. Among these is muscle tone. Without at least a minimal amount of muscle tone and control, the snowball wouldn't go anywhere. Another is the mental hand/eye coordination of the person throwing the snowball. Without such coordination, the snowball would be unable to reach its intended target, and therefore the same effect wouldn't be achieved. Yet another cause could be the existence of gravity. Without said gravity, the snowball wouldn't have the same trajectory and force behind its impact, and therefore wouldn't have the same effect.

As was stated before, I think background conditions are causes. The average person doesn't see them as such because they are looking at the direct cause and effect of whatever event.

~S. Vetterly

Jesse Steinberg said...

Steph- This sounds like a pretty plausible idea. You seem to be getting at a distinction between DIRECT causes and INDIRECT causes. Do you think there is such a distinction? What do you think it amounts to exactly? It's surely different from a temporal chain--like where A causes B and B causes C and so A causes C in a, I don't know what you might call it, a "temporally distal" sense. The presence of oxygen doesn't come BEFORE the striking of the match. There has to be oxygen when the match is struck (and thereafter). So, again, I'm wondering what you might have in mind in terms of this "direct vs. indirect cause" distinction that is implicit in your post.

Anonymous said...

This is really a great blog! Causality is a really fascinating topic.

What if you tried to lay out the conceptual framework for talking about causality in terms of the folowing components: a "causal agent," a "pathway of influence," a "mediator" (a link in a causal chain connecting the causal agent to the causal consequence), a "moderator" ((background conditions that facilitate or interrupt the causal pathway from the causal agent to the causal consequence), and a "causal consequence."

Given this, you can distinguish distal causes from proximal causes (as they fall in the causal pathway), and distinguish causal agents from moderators that influence causal relationships. So for an example of a moderator, some chemical reactions are facilitated in the presence of a catalyst. The catalyst is a moderator of the relationship beteen the cause (putting the substance in acid) and the effect (the substance disolving). The contact with the acid is the cause of the substance disolving. Note that the moderator (catalyst) is not caused by the dropping of the substance in acid.

I'll stop at this point to see if there are any problems with this.

Jesse Steinberg said...

I've got a guess about anon's identity... but I assume he/she is anon for a reason. Thanks for commenting!

This suggestion (moderators, mediators, etc.) is something that came up in class today. It may indeed be the beginnings of a way of distinguishing between causes and background conditions in a way that excludes the latter from counting as genuine causes. I'm hoping that one or more of the students in my class will write on this topic.

One of my students (Arpad) made a pretty strong case for being generous with the ascription of 'cause' and I think many would share the intuition that there are many causes of a given effect. The interesting question given the introduction of the concepts above is what differentiates mediators/moderators from causes and why shouldn't we also count those mediators/moderators as (also being) causes? It's not clear to me from what anon says above that this issue has really been solved. All we've done is replaced 'background conditions' with more precise language. But the issue still seems to remain even with this new terminology.

Anonymous said...

Well, consider the following classic example of a moderator. Suppose that the divorce of a child's parents generally causes more agressive behavior in boys as compared with same age girls. The cause of the agressive behavior (through some complicated pathway) is the divorce of the parents (and all that goes along with it). The sex of the child is not caused by the divorce; it is a moderator of the effect of the divorce on chldren. Conceptually, it is distinguishable from the causal agents at work in this scenario.

The effect of the divorce on the child is mediated by all that is involved in a divorcing parents scenario (e.g., arguments over custody, general hostility, impaired parenting, moving, changing schools and friends, excessive discipline, etc.). This, by the way, indicates that an effect may be due to the operation of a number of causal agents (all the mediators that are involved in a divorce), and scientists try to determine the relative contribution of each of these potentially causal agents in bringing about the result, i.e., increased agression. Some mediators are integral in the pathway from cause to effect, and the causal agent may only produce its effect through the mediators.

What do you all think?

Jesse Steinberg said...

Anon--The examples are really helpful in fleshing things out and trying to get clear about this.

So is your opinion that the mediators are causes? More precisely, are they one of the many causes of an effect? If so, then the picture of causation you've got is one that includes what some call "background conditions" as causes. This sounds strange to many, but (as we discovered in my class) many don't see anything wrong with this picture. One thing that we might think more about is why it we--as ascribers--don't say things like:

"The presence of gravity is (part of) what caused the bird to land on the tree."

We usually attribute the bird's desire to feed her young, her being tired, the fact that the tree provides shelter from a predator, etc. These are the "standard" sorts of things that might count as causes of the effect in question. Ordinary people wouldn't even think of gravity as part of the list of causes. Does this show anything about what REALLY count as causes?

----An aside, any time philosophers (or anyone) use the word 'really' like this, an eyebrow should be raised. I think such a move usually shows that there's a confusion or something to be very careful about that has to do with a departure from what might be called "common-sense."---

But this isn't to say that the question I just raised is an illegitimate one. I'm genuinely curious about how ordinary-person causal ascriptions should be weighed in evaluating what count as causes.

Anonymous said...

Here are some definitions from a paper I am working on with Jesse and our friend Chris Layne. These definitions of key terms involving causality may be helpful to further the discussion.

• A partial or contributory cause: a causal agent that plays a role in the cause/effect relationship
• A necessary cause: a causal agent without which the effect cannot occur
• A sufficient cause: a causal agent that guarantees that the effect will occur
• A sole cause: a causal agent that is both necessary and sufficient for the effect to occur
• A conjunctive cause: a causal agent that must be jointly present with one or more other causal agents for the effect to be produced
• A disjunctive cause: a causal agent among a number of other causal agents, any of which may produce the same effect
• A reciprocal causal relationship: occurs when two things are both causes and effects of one another
• A causal chain or multi-step causation: occurs when there are a number of successive causal agents that produce an effect
• A pure moderator: a variable that neither exerts a causal influence on the effect, nor is itself an effect of the causal agent, but rather interacts with the causal agent such that the effect varies as a function of the moderator
• A mediator: a variable that functions as a link in a causal chain connecting a causally prior factor to a causally subsequent factor
• A fully mediated relationship: occurs when the effect of a causal agent is fully transmitted via the mediator
• A partially mediated relationship: occurs when the effect of a causal agent is transmitted both by a direct pathway and via the mediator

As there is often confusion about the difference between a risk marker and a causal risk factor, s risk marker is a variable associated with an increased risk of an outcome and is, by definition, correlational in nature. As such, risk markers are not necessarily causal and are evaluated by comparison of the risk of those exposed to the potential risk factor with the risk of those not exposed. Thus, the following definitions should be stressed:
• A risk marker is a variable that is associated with an outcome or effect, where direct alteration of the risk marker does not necessarily alter the risk of the outcome.
• A causal risk factor is a variable that is causally related to the outcome and is an appropriate target for prevention or intervention.

Hope this is useful.

Jesse Steinberg said...

As anon knows, I think these definitions can be quite helpful in getting clear about causal processes.

But, at least in what's been discussed thus far, I don't see any solution to the problem at hand. Maybe there's something here, but I'll leave it for those reading this blog to chime in.

Michael Spong said...

The background conditions are causes because with out them the desired effect would not result. Thus; the "cause and effect theory" comes into play. People dont refer to causes when they state something because most causes are just taken for granite, without really thinking about the actual conditions. Therefor, conditions and causes are one in the same.

Corey Gibson said...

I do not believe there is any important metaphysical difference between causes and background conditions. Her striking the match, not having any wind and the presence of oxygen are all causes or reasons why the match became ignited. All of those examples are a producer of the effect which is the match being ignited. This may all seem odd but there could be many causes to this situation.

Anonymous said...

A cause sufficiently complements necessary background conditions to produce an effect.