[lutusp]

>> In science, the explanation is both the cart and the horse.

> I'm sorry, but no. Observation comes first, then hypotheses, then prediction, then verification.

The corpus of scientific theory is a set of tested, falsifiable explanations. Legitimate sciences don't rely on mere descriptions, even well-tested ones.

But let's take your claim and test it scientifically -- let's assume that we don't need explanations, we can get by with your stated criteria: observations, "hypotheses, then prediction, then verification." Here goes:

Let's say I'm a doctor and I've created a revolutionary cure for the common cold. My cure is to shake a dried gourd over the cold sufferer until he gets better. The cure might take a week, but it always works. My method is repeatable and perfectly reliable, and I've published my cure in a refereed scientific journal (there are now any number of phony refereed scientific journals). And, because (in this thought experiment) science can get along without defining theories, I'm under no obligation to try to explain my cure, or consider alternative explanations for my breakthrough — I only have to describe it, just like a psychologist.

Because I've cured the common cold, and because I've met all the requirements that psychology recognizes for science, I deserve a Nobel Prize. Yes or no?

Ask yourself what's wrong with this picture, and notice that the same thing is wrong with psychology — all description, no explanation, no established principles on which different psychologists agree, no effort to build consensus, and no unifying theories.

> So, we have observations w/out a coherent, compelling, or generally agreed-to theory. If and when a successful theory is developed, it will predict observations to date and predict more effects not yet observed or observed and ignored. Well, that sounds like a pretty exciting field of science, actually.

Yes -- and shaking a dried gourd can cure the common cold.

[pradocchia]

To continue your thought experiment, another researcher designs a new, double-blind trial of gourds, and discovers that while gourds per se have no effect, the placebo effect is statistically significant.

There is no explanation for WHY expectations would effect outcomes. In fact, any explanation at this stage would be provisional and highly suspect. And yet, there it is. What are we to do, ignore this phenomenon as if it didn't exist? Question the statistical ability of the researcher, and all other researchers who document a placebo effect?

And lo, the messenger. Ready, aim, fire.

[lutusp]

> To continue your thought experiment, another researcher designs a new, double-blind trial of gourds, and discovers that while gourds per se have no effect, the placebo effect is statistically significant.

Without a search for causes, for explanations, even the placebo effect is routinely disregarded. For example, it has been recently discovered that all psychological therapies are equally efficacious. Until now, the assumption was that Cognitive-Behavioral Therapy was superior to others, but that's been disproven. But, even though all therapies produce the same outcome, no one in psychology seems willing to consider the idea that it's all placebo effect.

> In fact, any explanation at this stage would be provisional and highly suspect.

Except the one that Occam's razor suggests, the default assumption under these circumstances: placebo effect. Or, perhaps better, the non-explanation suggested by the null hypothesis -- nothing meaningful has been measured and no conclusions can be drawn, which I think is your point.

> What are we to do, ignore this phenomenon as if it didn't exist?

No, but as scientists, we would do well to avoid drawing any conclusions not supported by rigorous experiment -- including the responsibility to propose and then test a theory about what's been observed.

[md224]

I'm not really sure what you guys are arguing about... obviously observations and descriptions are an important first step toward developing scientific theories. Lutusp seems to be complaining about people who feel descriptions are sufficient and make no effort to discern the underlying mechanism. Is there really disagreement on this point?

If there is, then maybe this is just a debate about Instrumentalism:

http://en.wikipedia.org/wiki/Instrumentalism

[lutusp]

> Lutusp seems to be complaining about people who feel descriptions are sufficient and make no effort to discern the underlying mechanism. Is there really disagreement on this point?

Yes, among psychologists, who insist that explanations aren't necessary, that it's science even if no one tries to identify a cause for the effect being measured. But this assumption is now under serious challenge, as more and more emphasis is being placed on a search for causes, to the degree that the director of the NIMH recently ruled that the DSM (psychology's "bible") will no longer be accepted as a source for science (it will remain as a diagnostic guide).

The practical meaning of this change is that researchers who apply for funding through the NIMH will need to avoid using the DSM's symptomatic categories as a basis for research -- they instead need to express their proposals in more scientific terms, in terms of causes, not just effects. In other words, explanations, not just descriptions.