I'm surprised that it's your experience that "many scientists [still believe dopamine] to be a kind of 'pleasure chemical'". While I'm certainly not a scientist, I've done quite a bit of cursory reading about the role of dopamine in habit forming, addiction, and learning. I've also read about it a little in a neurobiology textbook I own. And I never got the impression that any scientists think that it causes euphoric feelings, only that its presence was strongly correlated with those and related feelings. In fact, the explanation that there's another common cause to both raised dopamine levels and euphoric feelings seems to me, on the face of it, to be much more plausible, especially considering the role of dopamine in mundane habit forming and memory.

the effect of drugs that directly activate dopamine receptors is not euphoric in humans

Can you link to this? It seems very significant. Does it mean that the compulsive self- stimulation of rats with electrodes planted in their nucleus accumbens occurs without euphoria? (the nucleus accumbens is dopamine- activated). Are they merely becoming obsessive- compulsives, without joy in what they do?

In fact, the explanation that there's another common cause to both raised dopamine levels and euphoric feelings seems to me, on the face of it, to be much more plausible, especially considering the role of dopamine in mundane habit forming and memory.

I don't know what you mean by "mundane" habit forming and memory. Do you mean learning that does not involve reward associations? In any case, there is a qualification I forgot to make: reward has been tied to mesolimbic dopamine in particular, not dopamine just anywhere in the brain. There are other dopamine pathways that specialize, for instance, in production of motor patterns, expaining the deficits in Parkinson's disease, which leads into my reply to dg's comment.

Direct dopamine agonists such as pergolide, bromocriptine, and ropinirole are used as treatments for Parkinson's disease, and have not been associated with recreational use in all the time they have been known. A good article to read if you are interested is one by Kent Berridge and Terry Robinson from 1998. Berridge is one of the leading scientists trying to prove that dopamine is not associated with reward per se. In contrast, he believes that the "liking" or pleasure produced by rewards is mediated by opioid peptides. He bases this upon experiments in which rats showed a "happy facial expression" toward a sweet taste and a "disgusted" expression toward a bitter taste. Destruction of dopamine neurons did not block this, but administration of opioids incleased the "pleasant" response to the sweet taste. The problem I have with this assay is that it measures how the drug affects the "liking" of a secondary, food reward rather than the subjective effect of the drug alone. This extra degree of separation could make a huge difference, an idea that is supported by the fact that the known euphoria-producing drug amphetamine doesn't enhance the sweet taste response and so tests "negative" (Berridge himself says this here).
The opioid theory does not explain why dopamine reuptake inhibitors often produce euphoria.

As for the electrical stimulation, the nucleus accumbens contains neurons that use many diffeent neurotransmitters, including opioid peptides, in addition to dopamine.

A little off topic, but I've never heard of an explanation of why cocaine and amphetamines have such similiar effects if amphetamines make dopaminergic transmission independent of pre-synaptic firing but cocaine does not assuming that is really true.

Regarding the above post: The only difference between amphetamine and dopamine is a pair of hydroxyl groups on the 4th and 5th carbons of the phenyl group. Which gives amphetamine the ability to dock with dopamine receptors thus bypassing the Pre-synaptic neuron.
From random reading of articles I've gathered that one of the more immediate consequences of increased serotonin release is to facilitate the firing of the post-synaptic neuron. Additionally I wanna say that serotonin is modulated more by the extremity of a reward as opposed to the presence or expected presence of a reward as you see in some dopaminergic regions like the Striatum.
T