Report R2.5 Probability, Induction, and the Logic of Research (Part I) Seminar Leader: D. P. Dash, XIMB dpdash[at]ximb.ac.in Part 1 Varied images of Science (or research) exist in academic discussions. Science could be examined as a body of knowledge, as a practice or as a form of interaction among independent observers/actors. Numerous attempts have been made to demarcate science from non-science or pseudo-science. One very commonly proposed distinction is between Science and Art. The former is normally understood to be a body of objective knowledge whereas the latter is concerned with subjectivity, creativity, values, actions, application of knowledge, use of tools, etc. However, such a demarcation process breaks down when we see that all these elements of art are also present in science or scientific practice! Therefore, the issue of demarcation needs other types of analysis. There was a discussion about employing notions of activity, aim, and method to segregate research from non-research. However, these too have not altogether been successful, as scholars have raised a number of difficulties with these. Lakatos opines that research projects cannot be differentiated and that only research programs (consisting of a series of projects) are amenable for demarcation. Some scholars have expanded this further and concluded that what we should examine are research traditions that consist of a series of programs. Feyerabend takes an extreme view on this and states that all discussions on demarcation are futile as differentiating research from something that is not, is impossible. De Zeeuw prefers to leave the whole question to the requirements of the particular area of research being studied and entrusts the researchers with the responsibility of constructing suitable demarcation criteria that enable them to conduct meaningful research in their emerging research contexts. Comparison and grouping of observations are central to research and criteria employed for this are important devices that could differentiate research from non-research. A true scientific class can be developed only if stringent criteria that achieve observational closure are employed. Social acceptability or endorsement by an authority is not sufficient for this. A scientific class and its members are inter-dependent, but can prevail without support from an external authority. This is what Hacking refers to as the "evidence provided by things." Often it is easier to study changes of a class over time than to define and establish the class a priori. The "Contestablity" or "Falsifiability" of research is another characteristic that sets it apart from other activities. A genuine researcher tends to assert with Karl R. Popper that "all theories are hypotheses and that all may be overthrown". Part 2 Ian Hacking identifies two aspects for probability. One is an epistemological attempt at examining the credibility of propositions in the light of judgment or evidence. Many modern philosophers view it as a relation between a hypothesis and the evidence for it. In other words, probability operates at the stage where a sign is turned into evidence. The other is the statistical aspect concerned with the laws of chance and the tendency of certain events to produce stable frequencies. The Probability that an event E occurs is usually expressed as P(E) = [Number of times E occurs / Total number of outcomes] when the number of trials tends to infinity. This notion is also expressed as the long-run relative frequency. In some cases, as the total number of trials approaches infinity, P(E) tends to converge to a particular value. This tendency to converge could be viewed as a property of an infinite sequence. Can we consider the "sequence" itself as an object with stable properties? "Infinite sequence" as a potential scientific object seems to fulfill the conditions necessary for a genuine scientific class that we discussed earlier. Thus the class of "large sequence" becomes the object, and consists of a number of instances of specific large sequences. The tendency to converge is a property of this scientific object. Reference Hacking, Ian. (1975). The emergence of probability: A philosophical study of early ideas about probability, induction and statistical inference. London: Cambridge University Press. (Chapters 1-4) Reported by Jacob D. Vakkayil, with inputs from D. P. Dash.
Xavier Institute of Management, Xavier Square, Bhubaneswar 751013, India Research World (ISSN 0974-2379) http://www1.ximb.ac.in/RW.nsf/pages/Home |
