From BiodivBorneo09

Jump to: navigation, search
Science is a social activity and science progresses by the interpretation of accumulated observations. Interpretation is always subject to debate, and scientific papers are essentially statements of arguments.
Scientific argumentation is ultimately about:
  • developing, warranting, and communicating a persuasive argument in terms of the processes and criteria valued in science
  • constructing, critiquing, and communicating sound and valid arguments in terms of the connections between and among the evidence and theoretical ideas[1]
Perhaps the most radical statement of the social grounds of scientific knowledge appears in Alan G.Gross "The Rhetoric of Science." Cambridge: Harvard University Press, 1990. Gross holds that science is rhetorical "without remainder," meaning that scientific knowledge itself cannot be seen as an idealized ground of knowledge. Scientific knowledge is produced rhetorically, meaning that it has special epistemic authority only insofar as its communal methods of verification are trustworthy. This thinking represents an almost complete rejection of the foundationalism on which argumentation was first based.[2]
Typically an argument has an internal structure, comprising of the following
  1. a set of assumptions or premises
  2. a method of reasoning or deduction and
  3. a conclusion or point.
An argument must have at least one premise and one conclusion.[2]

The usual method of reasoning in science is 'informal logic,' which is basically logical common sense.

Analyzing arguments

Figure 1. An argument diagrammed in Carneades[3]

Arguing well, i.e., concisely and robustly, is surprisingly difficult, but is a skill that is vital to develop, since it underlies both science and our general ability to think clearly and not be fooled by rhetorical tricks and fallacies. There exist several software applications that help clarify arguments by breaking them down, usually with visual means. Two good (free) examples are Araucaria and Carneades (see Fig. 1).

We will not be asking you to use diagramming software, although you are welcome to use it if you want. We are, however, going to ask you to analyze (and refine) the argument behind each report you write. The key elements of analyzing arguments are[4]:

  • Argument graphs are hierarchical, with one conclusion and one or several, possibly nested, premises (when several premises descend from a single conclusion or premise they are referred to as linked arguments). Please lay out the premises as hierarchically arrange bullets. Premises can be conclusions of sub-premises, and sub-premises can be conclusions of sub-sub-premises, etc.
  • Your conclusion (or terminal statement), is the main idea that the paper is written to convince the reader of.
  • Your premises support that conclusion in various ways. Premises can be either 'pro' (the default) their local conclusion, or 'con.' If 'con' an additional statement is needed to indicate why the premise can be discounted. Add a qualifier (usually, sometimes, etc.) if appropriate. Premises sentences usually begin with "because." Premises can be optionally additionally labelled as:
    • Accepted concepts from literature (equivalent to backing)
    • Your own datum (your statistically-tested findings and other observations)
    • A 'local' assumption, needed for the logic of your argument, but not definitively established (conceptually related to missing premises).


  • Harry is a British citizen (conclusion)
    • A man born in Bermuda will generally be a British citizen (premise)
      • Civil code article 123 (sub-premise, literature)
    • Harry was born in Bermuda (datum)
    • Harry has not been naturalized into another nationality (assumption)
    • Joe reported that Harry was Russian (datum, con), but we do not trust Joe (discounted)


  1. Clark, D. et al. "Scaffolding scientific argumentation between multiple students in online learning environments to support the development of 21st century skills." Presentation slides from internet.
  2. 2.0 2.1 Wikipedia. 2009. "Argumentation theory"
  3. Gordon, T. F. 2007. "Visualizing Carneades Argument Graphs"
  4. Numerous schemas have been proposed to help dissect arguments, the most well-known being that of Toulmin (1958, "The Uses of Argument." Cambridge University Press). The one proposed here is a hybrid, optimized for teaching science. Since it represents the arguments in a single paper, we will not introduce the functionality of assigning arguments to different people.
Personal tools