The 'Descriptive' Curse

Getty Images "The work is basically sound but suffers from being too descriptive." Anyone who has served on grant review panels, participated in promotion and tenure decisions, or made and received editorial decisions, has heard this refrain or its equivalent and knows full well its kiss-of-death implication. As applied above, there is the unstated but intended linkage of the term "descriptive" with a nonmechanistic, non-hypothesis-driven approach to the work under evaluation. It's a comment that has both puzzled and irritated me. The Oxford English Dictionary defines descriptive as "characterized by description," and description as "a statement which describes, sets forth or portrays; a graphic or detailed account of a person, thing, scene, or event; i.e., a verbal or pictorial representation." Nowhere does the term imply an absence of mechanistic thinking or a lack of imagination. Quite the contrary, it is the essence of all physical and biological science. A careful and accurate description provides a true-to-life and faithful representation of a scientific observation. In fact, I cannot imagine any field in the physical or life sciences that does not depend on solid observations providing the foundation for all subsequent conceptualization in that discipline. But descriptive input does not end when conceptualization begins; tests of any hypothesis must contain precise observational analyses. Thus, in my view it is naïve and shortsighted to artificially separate hypothesis-driven research from descriptive studies; both must be integrated in productive experimental science. More to the point, however, is the unstated presumption that morphological studies of a cell, tissue, organ, or a whole organism are less creative, conceptual, hypothesis-driven, or experimental than deriving a gene sequence, a protein crystal structure, a gene array scan, or the creation and analysis of a transgenic animal. These are all descriptive, differing only in their limits of resolution and the current fashions of scientific research. Unfortunately, the adjectives descriptive, dull, and nonmechanistic have become lumped negatively together in the minds and actions of too many colleagues, especially in the field of biomedical research. Some of the most profound insights in the biomedical sciences have been based on pure morphological analyses of subcellular and intercellular structure. Let me cite just two examples: the elegant studies of Farquhar and Palade,1 defining the structure of intercellular junctions in epithelia, and the studies of Heuser and Reese on synaptic vesicle recycling at the neuromuscular synapse.2 The former study literally created a field defining how cells link to one another, how nutrients move between cells, how intercellular relationships are disrupted during metastases, and how tissues are constructed during development. The latter work has provided the underpinning for subsequent mechanistic studies of all chemical synapses, in the peripheral and central nervous systems. Hypothetical considerations of quantal neurotransmitter release would involve an awful lot of hand-waving without this beautiful electron microscopy. What made these studies so valuable was not the presence or absence of descriptive analyses; it was the enormous insight and predictive quality of that research and how it transformed all subsequent thinking in the fields of cell biology and physiology. Two of the most generously supported disciplines of our era are the Human Genome Project and protein structural biology. Aren't these two fields--and this is not meant pejoratively--the epitome of descriptive research? Is that work less descriptive than a careful analysis of cell morphology, relating the structure of its organelles to physiology, or to transitions that accompany embryonic development or disease? What should be clear is that good descriptive research is every bit as valuable and worth supporting as genetics, gene therapy, molecular medicine, or the more classical disciplines of physiology, biochemistry, immunology, pathology, pharmacology, microbiology, and anatomy. In fact, all these disciplines depend on careful and insightful descriptive research to both inform and enrich their fields. One of the strongest concepts of modern cell biology has been the recognition that structure and function are tightly linked. To be sure, knowledge of structure will permit a good cell biologist to make cogent predictions of function. It is this same thinking that underlies the hope of genome sequencers, that knowledge of protein sequence (primary structure) will soon permit realistic predictions of protein function. It is obvious that large conceptual and data gaps exist between gene and protein sequences, on the one hand, and the organization, physiology, or development of cells, tissues, organisms, and populations on the other. Extrapolations from the gene or protein crystal structures to higher levels of organization and function are not straightforward, nor easily solved. Descriptive research will be indispensable in bridging this gulf between genetics and "epigenetics." It is time to dispel the notion that descriptive research is, a priori, less conceptual, hypothesis-driven, or mechanistic than any other field of biomedical research. Of paramount importance to any reviewer is identifying the "originality, creativity, and predictive nature" of a research study, not damning it with the misnomer descriptive. Donald Fischman, MD, (fisch@med.cornell.edu) is the Harvey Klein Professor of biomedical sciences in the Department of Cell and Developmental Biology, Weill Medical College and Graduate School of Medical Sciences at Cornell University. References 1. M.G. Farquhar, G.E. Palade, "Junctional complexes in various epithelia," J Cell Biol, 17:375-412, 1963. 2. J.E. Heuser et al., "Synaptic vesicle exocytosis captured by quick freezing and correlated with quantal transmitter release," J Cell Biol, 81:275-300, 1979. function sendData() { document.frm.pathName.value = location.pathname; result = false if (document.frm.score[0].checked) result = true; if (document.frm.score[1].checked) result = true; if (document.frm.score[2].checked) result = true; if (document.frm.score[3].checked) result = true; if (document.frm.score[4].checked) result = true; if (!result) alert("Please select") return result; } .myradio{background-color : #94bee5} Please indicate on a 1 - 5 scale how strongly you would recommend this article to your colleagues? Not recommended 1 2 3 4 5 Highly recommended Please register your vote

The 'Descriptive' Curse

Interested in reading more?

Become a Member of