If this is more of an anthropological question, I'll repost it to /r/AskSocialScience.
Campbell draws a great deal from Carl Gustav Jung (1875-1961) who produced a series of enormous, detailed volumes. Jung's work is difficult to digest in part because of his encyclopedic approach to his material, but his was an earlier attempt to weave together a thoughtful overview of various myths from an international, pan-humanity point of view. And he is the source of the idea of a collective unconscious that pervades Campbell's work. Campbell did a great job in making Jung's scholarship accessible to the average reader, while adding his own layer of analysis and thought on the subject.
From a folklorist's point of view, there are problems with both Jung and Campbell. The following is an excerpt from my Introduction to Folklore. Campbell - and Jungian - enthusiasts will no doubt disagree with my assessment. I don't mean to be overly critical (I really respect Jung's work, and I appreciate what Campbell tried to achieve), I merely present at least one folklorist's point of view:
The popularity of one particular approach among non-folklorists warrants a digression. In the last part of the twentieth century, Joseph Campbell (1904-1987) created a great deal of interest in mythology and folklore with a series of publications on the subject. This was followed by a 1980s series of television interviews, which propelled Campbell to popularity with the general public, but not necessarily with all folklorists. To a certain extent, Campbell was relying on an older approach that Carl Gustav Jung (1875-1961) developed. Jung was a Swiss psychologist who studied with Sigmund Freud (1856-1939) but later broke with his mentor’s teachings to form his own approach to the study of the human mind. Jung developed the idea of the collective unconscious, maintaining in almost spiritual terms that all of humanity is linked by archetypes that existed in an unconscious common denominator. Ultimately, Jung implied that certain themes are woven into the fabric of the universe. This results, according to Jung, in a shared symbolic vocabulary which manifests in dreams, mythology, folklore, and literature.
Jungian psychology was extremely popular during the upheavals of the 1960s when people looked for mystical explanations of life to unify all existence. In spite of the faddish qualities of the late twentieth-century consumption of Jungian ideas, it is easy to regard Jung as an exceptional thinker with an extraordinary background of diverse reading. Campbell borrowed heavily from Jung, presenting many of these ideas in an easily consumable package that, in its turn, became something of a fad during the 1980s. Campbell drew not only on Jung, but also on Otto Rank’s 1932 publication, The Myth of the Birth of the Hero.
There are clearly many good ideas in this literature, but there are problems with the approach of Campbell, Jung, and Rank from the point of view of folklore studies. The first is that they tend to present the concept of tale types in mythology and folklore as though it were a new discovery. In other words, they ignore the highly developed bibliography that the discipline of folklore offers. The second, more serious problem is that this line scholarship makes no distinction between the core of a story and its culturally-specific or narrator-specific variants and variations. The Jungian-Campbell approach treats any variant of a story as an expression of the collective unconscious, regardless of whether its particular form is the product of an individual storyteller’s idiosyncrasies or of the cultural predilections of a region made irrelevant by traveling to the next valley. And with this process, all the other variants are ignored, including ones that may contradict the initial observation. This does not mean that there are no valuable insights in the work of Jung and Campbell. There are, of course, but folklorists regard their approach as removed from their own discipline and flawed, to a certain extent.
Dundes presented a similar critique of Freudian-based psychoanalysis of folktales. In his The Study of Folklore (1965), he wrote that “the analysis is usually based upon only one version…To comparative folklorists who are accustomed to examining hundreds of versions of a folktale or folksong before arriving at even a tentative conclusion, this apparent cavalier approach to folklore goes very much against the grain. How does the analyst know, for example, whether or not the particular version he is using is typical and representative.” (107) Dundes also pointed out that often the “variant” presented by the psychological analysis is from “a children’s literature anthology, rather than directly from oral tradition.”