Specialization is unfortunately an unavoidable fact of life as no one can ever hope to be an Aristotle of the 21st century. The emphasis on molecular and cellular biology is understandable but also unfortunate. The following passage from the book Natural History of Amphibians says it very well:
"In recent years teaching and research in natural history have declined in popularity, and greater focus is being given to cellular and molecular studies at many colleges and universities. On some campuses, natural history studies have never existed or have been eliminated completely. While the importance of studies at the level of molecules and cells is unquestioned, it is equally important to study life at the other end of the spectrum of biological organization—whole organisms and their interactions in nature. To paraphrase George Gaylord Simpson, knowing all there is to know about a lion's molecules and cells will not tell you why a lion roars."
As important as natural history and ecology are to the understanding of biological organisms, molecular data is indispensable to our understanding of evolutionary history. Hence I disagree with your categorization of molecular and genetic data as garbage. Often the real problem in molecular systematics is not the data, but the interpretation of data. Nevertheless there are some molecular studies which were improperly done and the results are thus unreliable. For example, there was one study which tried to examine relationships among the ratsnakes but which used the American descendants of a European ratsnake, namely Lampropeltis, as the outgroup. An outgroup species is chosen because it shares ancestral characters with members of an ingroup because of shared ancestry. By choosing an outgroup carefully, such shared ancestral characters can be found and these characters can then be used to build a phylogenetic tree. For example, by choosing the chimpanzee as an outgroup, a molecular systematists studying the evolution of living human populations can determine which genetic markers are ancient rather than recently evolved. This is not possible if an outgroup is not available. What the scientist did by choosing Lampropeltis as an outgroup was equivalent to choosing, say, a human being as an outgroup when studying the relationships among the chimpanzee, gorilla, and orangutan. The genetic markers shared by, say, a human and a chimp are of course not shared ancestral characters among all of the apes. In fact such characters are of course most likely recently evolved, after the human-chimp lineage had split from the gorialla-orangutan lineage. Therefore a tree built with data from such a study is likely to be unreliable. It is in actuality silly mistakes like these which are sometimes made by molecular systematists which present obstacles to their acceptance by other systematists and biologists.
Even when molecular systematists are able to avoid the pitfalls of silly mistakes, sometimes they do come up with conclusions that are contradictory to, or otherwise not supported by, their own data. An example would be the study of relationships among the rubber boa populations. According to the mtDNA data, all populations of rubber boas are more closely related to each other than they are to their closest relative, the rosy boa. The authors determined that the southern rubber boa is a different species even though there is nothing in the mtDNA data to support that conclusion. In fact, even if we use Hennig’s criterion to delimit taxa, an approach that is favored by at least some of the authors of the paper on the rubber boa, the entire complex of boas can be considered a single species, since they are all descended from a single ancestor, and no descendant population of that common ancestor has been left out. The entire rubber boa complex is a holophyletic group. And of course breeding experiments conducted by Richard Hoyer have shown that there is no premating or postmating isolation between the southern rubber boa and the northern populations. Yet the southern rubber boa was named a different species on the basis of several unreliable morphological characters by these authors. Their conclusion of the status of the rubber boa really has nothing to do with the genetic data. It is thus DNA studies like those on the ratsnakes and rubber boas which give the opponents of molecular data (e.g. systematists who rely on morphology) fodder for their arguments that DNA studies are unreliable.
When correctly done and carefully interpreted, DNA studies simply are indispensible. In many cases, DNA studies confirm conclusions made by morphologists. For example, morphologists had concluded that Lampropeltis is a descendant of Elaphe, and this is confirmed by mtDNA data. Genetic studies have shown that whales are in fact descendants of a member of the Mammalian order Artiodactyla. The closest relative of whales are in fact the hippo, the pig and cows according to DNA data. Sure enough recently discovered fossil evidence shows that the leg bones of whales contain the characteristic found only in the leg bones of artiodactyls. Whales are descended from an artiodactyl!
In other cases, molecular studies have revealed convergent evolution among closely related species (for example within the same genus), which may well be impossible to prove otherwise and something that few would suspect in the absence of molecular data.
In conclusion, while I agree with you and with Robert C. Stebbins that there is currently an overemphasis on molecular data, I also agree with Stebbins that molecular data is useful and important. Some of the molecular studies are in fact garbage, and some of the conclusions from the genetic studies are also garbage, even if the data is not. But I think you mean that there is a large number of genetic studies that are garbage. I would agree with that. I hope you are not stating that most or all of the studies based on molecules are garbage because it is simply not the case.
