A serious problem is the idea that applying the concept of “disease” determines the appropriate response.
In ancient times, if you got leprosy, smallpox, whatever, you were thought of as having been blighted by Og, with which came the implication that you’d done something to deserve it. It was therefore legitimate (within that worldview) to condemn those with what we would now think of as diseases, or (alternatively and apparently arbitrarily) excuse them as being bewitched. Thus the religious model of affliction aetiology governed the social response to those afflicted.
This religious model of affliction was supplanted by the disease model, particularly once pathogens were discovered. If, instead of being thought of as being blighted by God, you simply had the bad luck to be struck randomly by a pathogen such a smallpox, then it was hardly your fault. You were entitled to sympathy and compassionate care. The response - in this case, compassion - was once again dictated by the aetiological model.
This history was played out markedly in cases of mental illness - the famous “Malleus Maleficarum” was intended as a text on how to identify demonic possession, but is now recognised as one of the first organised descriptions of the manifestations of mental illnesses.
But not all “diseases” have an aetiology as morally simple as those involving random pathogens. Some involve questions of choice, and those which involve questions of choice involve questions of degree of capacity to choose. To the extent that choice is possible, moral judgment is possible. It is thus not a simple matter of allocating to a condition the descriptor of “disease”, and then treating that descriptor as determining for all purposes the appropriate moral response.
An example. Schizophrenics in many cases have no meaningful control over their behaviour, and evidence suggests the source of the disease is in biochemistry and genetics. That the source of the illness is out of the control of the individual is a strong indicator that a compassionate response is appropriate. That the patient’s behaviours are beyond his control is also a strong indicator that a compassionate response is warranted.
But it is not always the case that the behaviour of a person with schizophrenia is beyond their control. Control may be impaired, but not destroyed. One commonly sees this in cases where a person with schizophrenia commits an offence outside the ambit of their delusions or hallucinations, or does something which indicates that they were aware of the moral wrongness of their behaviour or could partially control it (like waiting for an opportune moment to commit an offence, or lying or taking other steps to cover it up.) To the extent that choice remains, so does moral culpability. Diminished responsibility does not mean no responsibility.
Similarly with alcoholics. The argument that a feature of the disease is a tendency to deny the existence of disease and to deny the need for treatment can be overstated if it is expressed in terms that are too absolute. The argument overlooks the fact that a wide range of alcoholics whose affliction covers the spectrum of severity of addiction can, and do, teach themselves (often with help) to refrain from consumption notwithstanding that feature of the disease.
There are many conditions (I choose a relatively neutral term to avoid the loaded term “disease”) where a person retains the power of choice, albeit in some ways impaired. Gamblers, the morbidly obese, and substance abusers are common examples.
People being people, there is self-serving tendency sometimes to overclaim the benefits of compassion which they assert should to flow them because of their “disease”. Thus, the woman on Dr Phil and Mitch Hedberg (who was an addict) had an interest in pretending that it was all as simple as drawing a straight line between the label “disease” and the conclusion that they deserved unalloyed sympathy. The analogy between alcoholism and HIV or lupus is imperfect.
But there is something of an industry in trying to identify some patterns of behaviour as new “diseases” in the hope that sympathy rather than condemnation will automatically follow. Hence, sex addiction as the new alcoholism for the Tiger Woods of the world. But a sex addict is not beyond moral and legal judgment should he cheat on his wife or rape someone, nor is a gambling addict should he steal from his employer. The reason is that the power of choice is retained.
It is also important to distinguish between a disorder and a disease. These terms are imperfect as well for what I want to discuss, but they will do for the present. Upthread, Chief Pedant asked the rhetorical question, "Why is a mugger who kills impulsively not equally a “victim” of his “disease”? BigT responded with “They generally are. Almost every psychologist I have ever met would consider this a mental disorder.”
The DSM IV makes it clear that its definitions of disorders carry no implication of moral or criminal culpability. Most of the personality disorders are defined in large measure operationally, by reference to the disturbance the disorder causes to the person. They are commonly conceived of as descriptions of a person’s personality rather than as a disease entity. A psychopath has an anti-social personality disorder. That does not mean he cannot make choices, nor does it mean he is excused liability for his decisions.
All the serial killers are probably psychopaths. We imprison them not on the purely pragmatic theory that we have to protect ourselves somehow - we could put them in mental institutions for that. We imprison them because they do not qualify as having a disease, with all the connotations of sympathy and compassion I described above. They are not deprived of choice.
And the same goes for sexual disorders, particular paedophilia.
There is no doubt paedophiles have a strong compulsion to have sex with children, but they can resist the temptation, just as married people can resist the temptation to cheat with the uber hottie down the corridor at work.
Labelling something a “disorder” tells us little or nothing about how we should respond to someone displaying the signs of it.
So to answer the OP, my impression of the 1930s and 40s was that alcoholics were treated with contempt. Their condition was treated as entirely a character failing. Now, partly through the efforts of AA and partly through academic research, there is less condemnation, especially of those who at least make an effort. How hard it is to recover is better understood generally, although it seems also to be understood that there remains a component of choice in the condition. The remaining capacity for choice is not really capable of objective measurement either generally or in a particular case, so different people attach different weights to their judgment of how much the affliction is a consequence of choice. Hence, there is no clear answer to the question, beyond a sense of general improvement in compassion towards alcoholics.