In a weak form pragmatism is simply a theory about justification. The pragmatists claim (claimed) that the best justification for a claim or idea was whether it worked or didn’t. A claim that doesn’t have consequences can thus neither be justified or unjustified, not because it can’t be tested, but because it just doesn’t make a difference. “There can be no difference that doesn’t make a difference,” in the words of William James. Of course in this weak form pragmatism is not a claim that anyone is going to bother to dispute. Every theory about justification incorporates in some form or other the idea that the consequences of a claim matter when evaluating the claim. Pragmatism may call attention to that fact, but does not really stand out as a theory by virtue of it alone.
Where pragmatism is different from other theories about justification, and where it goes bad, is by taking this claim to extremes. Pragmatism claims that the only justification a claim can have is by “working”, by being assimilated, validated, corroborated, and verified (again in the words of William James). And pragmatism claims that there is no more to truth than being justified in this way. Unfortunately for such a strong version of pragmatism extending it in either of these ways makes the position self-defeating. Consider first the idea that justification consists only in a claim “working”. Suppose that a young scientist is attempting to formulate a new hypothesis about gravity. A number of ideas probably present themselves. According to pragmatism all of these ideas are at this point equally justified; since none have been put to the test none can be said to work better or worse than any of the others. But clearly this is not actually the case, certain hypotheses are already more justified than others by extrapolation from past theories. Hypothesis involving gremlins, fairies, or anything other than simple and unintelligent components interacting with each other are unjustified. If our scientist really withheld judgment about each hypothesis until putting it to the test then they would have to test each hypothesis that presented itself to them, since they would have no way to pick the one most likely to be a successful explanation. So, on pragmatic grounds, we must reject the pragmatist theory of justification if it is to stand by itself, since it simply won’t work. (What will work is pragmatism plus some rules that allow deduction from what is already justified.)
Similarly, trying to define truth as what is pragmatically justified also undermines the theory. The idea that there is some fixed and objective truth is pragmatically useful; it keeps people busy trying to uncover it, and in the process they come up with better theories. If we took pragmatism to heart then the process of creating better theories would grind to a halt, because there is little reason to look for another theory when you already have the truth. Thus pragmatism doesn’t work as a theory about truth, and so, by its own standards, would be false. So strong pragmatism rejects itself and endorses the competing view of truth.
Strong pragmatism is attractive to some because it promises to resolve a large number of skeptical dilemmas at once, since pragmatism labels nearly everything we already believe as true. But skeptical dilemmas only exist because we have high standards regarding truth. Thus getting rid of them can only be accomplished by effectively lowering our standards regarding truth, which pragmatism does. And lowering our standards regarding truth is not something to be happy about. It is only because we have high standards that we make intellectual progress or revise our beliefs. Consider the Copernican revolution. Scientists abandoned the idea that the earth was the center of the universe not because it worked better (at the time you could make better predictions with epicycles), fitted better with their other beliefs (many were religious), or was believed by the majority of people (it wasn’t). No, they abandoned the idea that the Earth was the center of the universe because the heliocentric theory seemed more likely to be true. But if pragmatism had been the prevailing opinion heliocentrism would never have caught on, because pragmatically the geocentric view was better. Even after the heliocentric theory was making better predictions it was still possible to preserve a geocentric world-view, by holding the Earth to be stationary, letting the sun revolve around it, and then letting the rest of the planets revolve around the sun. So, under pragmatism, there would have been no way to convince people to abandon the geocentric theory for the heliocentric one, no matter how good the heliocentric theory got. And that’s ugly; pragmatism and theories like it effectively prevent progress by making it effectively impossible to argue for the superiority of one theory over another as long as the entrenched theory “works” to some extent.