Connecting rhetoric of science, biomedical ontologies, and scientific writing - Part I - Relevance

Scientific writing is arguably the central form of expressing and persuading peers about the validity of scientific arguments, models, and theories. Currently, scientific writing in Biomedical Research is primarily expressed through peer-reviewed articles, although scientific books have played a major role in the past and the future will likely be shaped by other forms of communication. Independent from its form of expression, scientific writing still convenes the primary way by which researchers can evaluate and criticize the premises, methods, analyses and conclusions of their peers regarding a certain research topic. For example, very often the merit of scientific ideas is discussed having a scientific manuscript standing for the scientific idea itself. Examples abound in all areas of research, and one of the classics is the article "A Structure for Deoxyribose Nucleic Acid" by Watson and Crick in 1953, which resulted in the 1962 Nobel Prize. In more extreme examples, even the entire work of an individual researcher might be identified with a certain piece of writing, as was the case of John Nash's article "Equilibrium points in n-person games."

Despite its central role in the communication of scientific ideas, the way researchers are trained to persuade others regarding their ideas is still amateurish, frequently based on rules of thumb and lacking a formal, evidence-based analysis of former strategies that have been proved successful. This careless approach is not caused by a lack formal studies, since over the past few decades two distinct and largely non-overlapping disciplines have evolved to study the structure of scientific arguments and texts.

On one side, the field of Rhetoric of Science studies persuasion strategies and their associated text structure used to argue on behalf of a scientific argument. A classical example is the book by Ceccarelli (Ceccareli, 2001) evaluating how the researcher Dobzhansky used a series of rhetorical mechanisms (polysemy, prolepsis, metaphors, etc) to convince opposing naturalists and geneticists to accept his conciliating argument and finally work together. On the other side, the field of biomedical computational ontologies has dedicating part of its efforts to the description of the structure of written information within certain types of scientific articles. For example, Sim et cols. (Sim, 2004) have defined the formal structure of information elements constituting a randomized controlled trial, and the relationship among these elements. While one of the primary goals of biomedical ontologies in relation to scientific articles is to achieve computability of these texts (i.e., the ability to have scientific articles being interpreted by computers), rhetoric of science and biomedical ontologies share a common ground when it comes to the analysis of "text structure."

Text structure is the underlying skeleton used in a text. For example, in the Introduction section of a scientific manuscript, experienced researchers tend to use a sequence of arguments that usually follow a standard pattern:

1. Start by stating the significance of your topic so that peers can be convinced that your article is worth reading further
2. Once the significance is established, point to an information lag within the field that has not been addressed to date.
3. Substantiate the existence of this lag with a review of the literature to confirm that, despite other advances, this lag remains as an important and central problem.
4. Close the Introduction section by stating that the aim of your paper is to fill the lag you previously pointed to.

The structure outlined above could be used in virtually any content area. For example, one could use this structure to write the Introduction section of a paper focusing on problem in areas as diverse as Surgery, Physics, or Ecology. In other words, text structure and content are separate entities.

Where Rhetoric of Science and Biomedical Ontologies come together is that both use the concept of text structure, the former as a mans of persuasion, the latter as a way to formalize the data presented in scientific texts so that it can be manipulated by computers. The two fields are complimentary in that articles are written by humans to persuade others, a rhetoric activity, while the processing of ontologies by computers captures information produced by humans and ultimately deliver it back to humans themselves, a biomedical ontology activity.

This series of upcoming posts on the link among rhetoric of science, biomedical ontologies and scientific writing will attempt to bring some more clarity on their relationship elicited through and open discussion with members of the scientific community. Although the structure and sequence of these posts is not completely pre-defined, possible topics include the relationship between rhetoric of science and text structure of scientific articles, biomedical ontologies and text structure, how biomedical ontologies of scientific writing could be modified by the introduction of rhetoric of science concepts, areas of emerging work in the border between the two disciplines, and anomalies in biomedical ontologies that could be addressed by concepts from rhetoric. Finally, we will also discuss implications of adopting a rhetoric of science perspective to study scientific writing and their ontologies, as well as delineating potential areas for future work.

1. Leah Ceccarelli, Shaping Science with Rhetoric: The Cases of Dobzhansky, Schrodinger, and Wilson (University Of Chicago Press, 2001).