More on Gaining Reliable Knowledge: A Comment

Abstract

Some of shortcomings in way wildlife scientists view science and do research were discussed by Romesburg (1981). Romesburg (1981) suggested that there are methods that can by systematically followed to recognize patterns in nature (i.e., induction) and to formulate potential explanations for these patterns (i.e., retroduction). We argue that these methods need further consideration because mental processes involved are not really understood and are virtually out of our control. Also, his primary example of how to frame and test needs further development because hypotheses and field manipulations he proposed focused only on existence and nature of a pattern, and not on its explanation. We provide another example of hypothesis formulation and testing, but suggest that it is not technically feasible to use hypothetico-deductive (H-D) with most game species in field situations. We believe that explanations for many of patterns we see in game species will come indirectly from controlled experiments with testable animals that are ecologically similar to game species in question. J. WILDL. MANAGE. 53(4): 1172-1176 Some of major shortcomings in way wildlife scientists view science and do research were discussed by Romesburg (1981). The problems noted by Romesburg (1981) are not surprising because subject is complex and wildlife biologists often have restricted opportunities for exposure to abstract concepts of how science works or should work. What formal exposure does occur is often limited to intellectually sterile recipe for discovery offered in textbook treatments of the scientific method (Mayer 1984). However, there has been much discussion about how thinking and conduct of ecologists must be constrained and guided by an understanding of philosophy of science (Saarinen 1980, Salt 1983). There has been even more heated discussion over what that understanding is; of special concern is role of hypothesis formulation and testing in ecology (Strong et al. 1979, Willson 1981, Quinn and Dunham 1983, Simberloff 1983). Romesburg (1981) encouraged wildlife scientists to make greater efforts to formulate and test research rather than to merely draw conclusions from collections of general observational data. We agree with direction of Romesburg's (1981) message, but we wish to elaborate on some of details presented by Romesburg (1981) concerning formulation and testing of hypotheses. We also comment on appropriateness of example Romesburg (1981) provided of hypothesis framing and testing. We thank J. C. Tash for engaging us in study of philosophy of science, encouraging us to write this paper, and critically reviewing manuscript. N. S. Smith, R. Mazaika, and J. P. Hill also reviewed manuscript. GETTING TO HYPOTHESES Romesburg (1981) noted that wildlife scientists do a good job of making basic observations of nature, necessary starting point of science (Simpson 1964:88). Observational errors, optical illusions, and even hallucinations may distort our images of nature, but numerous observations from different points in space and time generally resolve such problems. Science, however, entails more than amassing observational reports. Romesburg (1981:294) reported that scientists may find patterns in their observations through a called induction. But nearly anyone, not just scientists, can recognize patterns (Hunt 1982). The ability to recognize patterns was certainly adaptive and enabled early man to note, for example, characteristics of sites likely to have food and water, areas likely to harbor predators, sky conditions commonly preceding storms, and signals to change of seasons. Modern man recognizes many different types of patterns, establishes their reality by using correlative statistics, and formalizes them into general rules of association. However, mental process by which we recognize patterns is not eally known, and we must wait for realiza ion of a pattern to arise in our consciousness. This is not to say that there is nothing that can be done to prepare for discovery, only that there is no of science that can be followed by

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 31 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.