Author: Bill Vallicella

I hazarded the following definition:

Belief B is superstitious =_df (i) B is or entails erroneous beliefs about the causal structure of the natural world; (ii) B makes reference to one or more supernatural agents; (iii) B involves a corruption or distortion of a genuine religious belief.

The conditions are supposed to be individually necessary and jointly sufficient.  Several people wrote in to question whether conditions (ii) and (iii) are necessary.  What about:  blowing on dice; avoiding walking under ladders; carrying a rabbit's foot, etc.

First off, these are not beliefs  but practices, and I had set myself the narrow task of defining 'superstitious belief.'   Second, it is not clear that  the people who engage in the aforementioned practices need have any underlying beliefs about the practices or their efficacy.  The gambler who blows on his dice before throwing them may simply be mimicking what he saw some other gambler do, a gambler he thought 'cool.'  Same goes for a liitle leaguer who crosses himself at the plate just because he saw some big boy do it.  Monkey see, money do.  The kid may have no idea what the gesture signifies.

But suppose our gambler really does believe that blowing on the dice will enhance their likelihood of coming up the way he wants.  Then (ii) and (iii) go unsatisfied.  But is the belief in question a counterexample to my definition?  Not unless it is a superstitious belief, which is what I deny!

"But doesn't the belief in question satisfy the dictionary definition?"  Yes, it does, but so what? I am not trying to give a lexical definition.  A lexical definition, or dictionary definition, aims to describe how a word or phrase is actually used at the present time within some linguistic community.  But if you think philosophical insight can be had by consulting dictionaries, then you commit what I call the Dictionary Fallacy.  People say the damndest things and use and misuse words in all sorts of ways riding roughshod over all sorts of distinctions.  When a misuse becomes widely accepted then it goes into the dictionary since dictionaries are descriptive not prescriptive.

But it occurred to me that there is a problem with my definition.  A belief can be superstitious even if it doesn't involve any erroneous beliefs about nature and her workings.  Consider again the plastic dashboard Jesus.  Suppose the motorist believes, not that the hunk of plastic has causal powers relevant to the prevention of automotive mishap, but that the divine person represented by the icon will be inclined to intervene in the natural world in prevention of mishap because he is being honored by the motorist.  Such a motorist could be a trained physicist who harbors no false beliefs about nature's workings.  (Divine intervention needn't involve any violation of natural laws.)  I want to say that that too is a case of superstition.  If it is, it is not captured by my definition. 

Superstition is a form of pseudo-religion, a degenerate or distorted form of religion.  But what exactly is it and how does it differ from genuine religion?  Let's start by asking what sorts of item are called superstitious.  There are (at least) superstitious beliefs, practices, and people.  Perhaps we should say that a person is superstitious if he habitually harbors superstitious beliefs and engages in superstitious practices.  Since practices are underpinned by beliefs, perhaps we can make some progress by trying to define 'superstitious belief.'

Go back to my example of the plastic dashboard Jesus icon.  The hunk of plastic has both physical and representational properties.  But properties of neither sort induce in the hunk any causal powers of the sort that are relevant to the prevention of automotive mishaps.  Now if the motorist  believes to the contrary, then he is superstitious — this seems to be an exceedingly clear paradigm case of superstition — and part of what makes him superstitious is that he harbors erroneous beliefs about the causal workings of nature.  So it seems that part of the definiens of 'superstitious belief' is

1. an erroneous belief about the casual structure of nature.

If (1) is a necessary condition of a belief's being superstitious, then the mere belief that God exists or that unembodied/disembodied souls exist is not superstitious.  Obviously, the belief that there are entities transcendent of nature needn't involve any false beliefs about nature.  We have to avoid the mistake of identifiying superstitious beliefs with beliefs about the supernatural.  That would be on a par with the mistake of thinking that religion just is superstition.

But (1), though necessary, is not sufficient.  For not every  erroneous belief about nature's workings is a superstitious belief. When I was a young child I got it into my head that my left arm had to be stronger than my right arm because, being right-handed, I used my right arm more and my left arm less with the result that the power of the left arm was preserved while the power of the right arm was reduced.  My childish belief was 'logical' in way, but empirically false.  Flexing a muscle is not like flexing a piece of metal.  The former typically strenghtens, the latter typically weakens.  But there was nothing superstitious about my false belief.    A second example is the gambler's fallacy which, though sometimes classified as a superstition, is not one by my lights.  So it looks as if we need to add a second necessary condition along the lines of

2. That makes reference to a supernatural agent.

Thus in the case of the dashboard Jesus what makes the belief superstitious is not the attribution to a hunk of plastic as a mere hunk of plastic of causal powers it cannot possess; it is the attribution of such powers to a hunk of plastic that is also iconic or representational, the item represented being a supernatural agent.  If the icon were melted down into a non-representational blob, then the superstitious motorist would presumably no longer consider it causally efficacious in warding off danger.

But now it appears that our two necessary conditions are not jointly sufficient.   I am assuming that superstition is a form, but not the only form, of pseudo-religion.  (Idolatry and blasphemy may be other forms.).  As a form of pseudo-religion, superstition is a degenerate or corrupt or distorted form of genuine religion. Now suppose our motorist is a member of a Satanic cult and has on his dashboard an icon that represents some demon or maybe the head honcho of demons, old Mephistopheles himself.    And suppose our satanist believes that the presence of that icon (made of the molded excrement of a sacrificed cat) will protect him from the dangers of the road.  Then both (1) and (2) will be satisfied without the satanist's belief being superstitious.  So I add a third necessary condition:

3. and involves a corruption or distortion of a genuine religious belief.

Example.  A kid makes the sign of the cross as he steps up to the plate in a baseball game.  If the kid believes that the gesture will increase the likelihood of his connecting with the ball, then he has an erroneous belief about natural causation.  But that is not enough to make his belief superstitious.  Nor is it enough if we add the  reference to a supernatural agent.  We need to add the third condition.    The genuine religious belief being distorted here is the belief that one's spiritual salvation depends on right relation to God, a right relation that can be secured only via the mediation of Jesus Christ. This genuine religious belief may be false but it is not superstitious: it does not involve any erroneous beliefs about the causal structure of nature.  The distortion consists in the invocation of Jesus and his self-sacrifice for a paltry mundane self-serving and ego-enhancing purpose having nothing to do with salvation.

This seems to do the trick.   My claim is that my three conditions are individually necessary and jointly sufficient for a belief's being superstitious.  Counterexamples anyone?

A reader comments:

You write that “Superstition in this first sense seems to involve a failure to understand the causal structure of the world or the laws of probability” and that it is a “necessary (but not sufficient) condition of a belief's being superstitious is that it entail one or more erroneous beliefs about the causal structure of nature.” I’m curious about what you think of the following case:

A young Christian claims to be harassed by demons. He experiences the harassment, he claims, only during the late hours of the night in his room. His pastor advises the kid to anoint with oil, in the shape of a cross, the door frame to his room (the idea being that such an anointing will “ward off” the demons). 

Do you think this is superstitious? After all, oil is in the relevant sense “material stuff” just as much as a plastic Jesus is. However, the oil is not intended to have a causal effect in nature, but in “the spiritual realm.” I think examples like this are common among religious people. It may not be hard to find one that intuitively counts as superstitious (as another example, consider how Catholic priests are often asked to bless houses or rooms before a Catholic takes up residence in them). What you think of these kinds of cases?

The more examples the better.  Yours is importantly different from the plastic Jesus example in which a power is imputed to a physical thing that it cannot have, the power to protect the vehicle and its occupants from a natural threat.  (Contrast this with the power a properly fastened seat belt has to prevent the driver from going through the windshield in the event of a crash.)  In your example there is imputed to a physical stuff, oil, the power to protect against a physical or spiritual threat emanating from a purely spiritual being.  Since this is a power that oil cannot have, whether applied in the shape of a cross or not, I would say that this type of practice and the underlying belief are superstitious as well.

It is worth noting, however, that a false belief can have a real effect.  Believing, albeit falsely, that he has done something efficacious to ward off demons, the kid may feel reassured and comforted.  The pastor's belief that the kid's daubing the door frame with oil will have a beneficial psychological effect on him is not superstitious.

I once knew a chess player who always wore the same ridiculous little hat, filthy and tattered, at tournaments.  This was his 'lucky hat.'  Donning it, he geared up for chessic combat.  This may or may not be a superstitious practice depending on the underlying belief.  If he believes that the mere donning of the hat directly influences the outcome of games, then the belief is superstitious, or at least bears one of the marks of a superstitious belief.  But there would be nothing superstitious about the belief that donning the hat puts him in a fighting frame of mind, which in turn does have a real effect on his play.

Or consider an airline pilot who suits up prior to a flight.  Donning his uniform, he steps into his role.  'Looking the part' he inspires confidence in himself and in his passengers.  This confidence has a slight but real effect on his performance in the cockpit.  So far, nothing superstitious.  Superstition would come into the picture if the pilot thought that the mere donning of the uniform enhances his skill set, that the insignia, say, have the power to confer upon him good judgment or motor control.

 This is a difficult topic.  Surprisingly little has been written on it by philosophers.  In the JSTOR database I found only four articles, three from the 1930s.  If we don't know what superstition is, then we won't know what genuine religion is either.

John Pepple has written an excellent post in which he sketches a religion free of superstitious elements, thereby showing that there is nothing in the nature of religion — assuming that religion has a nature — that requires that every religion be wholly or even in part superstitious.  Here is his sketch:

1. God exists.
2. Upon creating, God placed all sentient beings in heaven.
3. Some of us sinned and were sent to our universe for punishment.
4. There is no intervention by God in our universe, because that would interfere with the punishment.
5. After we die, we either regain heaven or are reincarnated.
6. We regain heaven not through worship of God but by good behavior, by treating other sentient beings right. In other words, we regain heaven by merit and not by grace.

As I suggested  in Religion and Superstition, the bare belief that there are supernatural beings is not superstitious.  Without essaying a logically impeccable definition of 'superstitious belief' (very difficult if not impossible), I would say that a necessary (but not sufficient) condition of a belief's being superstitious is that it entail one or more erroneous beliefs about the causal structure of nature.  I have seen Catholic baseball players  make the sign of the cross before stepping up to the plate.  That bit of (disgusting) behavior is evidence of a superstitious belief:  clearly the gesture in question has no tendency to raise the probability of connecting with the ball.  Or consider the plastic dashboard Jesus that I mentioned before.    The belief that the presence of this hunk of plastic will ward off automotive mishap is superstitious, and a person who occurrently or dispositionally has many beliefs like this is a superstitious person.

But what if the person believes, not that the piece of plastic will protect him, but that the purely spiritual person represented will protect him by intervening in nature?  That too is arguably superstitious, though not as egreuiously superstitious as the first belief.  One might argue like this:

a. The physical domain is causally closed.
b. The belief that Jesus will intervene in the workings of nature should one, say, have a blow-out is an erroneous belief about the physical domain.
Ergo
c. The belief in question is superstitious.

To make things hard for the religionist  suppose we just assume the causal closure of the physical domain:  every event in the physical universe that has a cause has a physical cause, and every effect of a physical cause is a physical event.  The idea is that no causal influence can enter or exit the physical domain.  That the physical domain is causally closed is neither obvious nor a principle of physics.  It is a philosophical thesis with all the rights, privileges, and debilities pertaining thereunto.

But even if causal closure is true, it doesn't rule out the existence of a wholly immaterial God who sustains the universe at every instant but never intervenes in its law-governed workings.  As far as Pepple and I can see there is nothing superstitious in the belief that such a God exists.  So there is nothing supersitious about Pepple's (1).

I read his (2) as the claim that God creates purely spiritual beings who exist in a purely spiritual domain.  Please note that sentience does not entail having physical sense organs.  For example literal visual seeing does not require the existence of physical eyes. In out-of-body experiences, subjects typically have visual experiences that are not routed through the standard-issue optical transducers in their heads.  And yet they literally (and arguably veridically) see physical things, e.g., the little bald spot on the top of a surgeon's head. 

Ad (3).  How do we get sent into this penal colony of a world?  We are born into it: the preexistent soul begins to inhabit an animal organism.  Soul in this sense is of course not an Aristotleian animating principle or a Thomistic anima forma corporis, but a Platonic soul.  But wouldn't the attaching of a pre-existent soul to an already living organism involve some violation of causal closure?  Not obviously.  But this is a deep question. (I now invoke the blogospheric privilege entailed by the 'Brevity is the soul of blog.')

Pepple's is a rather 'thin' religion but I think it illustrates nicely how religion and superstition can be decoupled.  For his is a belief system that counts as a religion but is clearly not superstitious.

What we need to make this really clear are definitions of 'religion' and 'superstition' ('pseudo-religion').  But definitions in this area are very difficult to come by.  And it may be that religion and superstition are both family-resemblance concepts that are insusceptible of rigorous definition in terms of necessary and sufficient conditions of application.