PSA2016: The 25th Biennial Meeting of the Philosophy of Science Association

Full Program »

Fitness Measures and Abstraction

Building on earlier work of one of the authors, we discuss the utility of and relationship between two different fitness measures: the expected number of offspring and the expected frequency of offspring. We argue that, if the probabilities underlying fitness values result purely from abstracting from selectively irrelevant causal features of an organism’s or type’s environment, this has important implications for both the utility of and the relationship between these two ways of quantifying fitness.

Based on the work of Gillespie (1977) and Beatty and Finsen (1987), it has been argued that fitness ought not be quantified as an organism’s or trait’s expected number of offspring (e.g., Brandon 1990; Sober 2001). Suppose there are two types in the population: As and Bs. What Gillespie and Beatty and Finsen show is that [(The expected number of As )÷(The expected number of As+The expected number of Bs)] will not correctly reflect the expected frequency of A offspring, and the latter is a more accurate way to quantify fitness.

In contrast, we argue that, if the probabilities underlying fitness values result purely from abstracting from selectively irrelevant causal features of an organism’s or type’s environment, then [(The expected number of As )÷(The expected number of As+The expected number of Bs)] is more accurate in quantifying what happens when selection is acting alone. Hence, [(The expected number of As )÷(The expected number of As+The expected number of Bs)] is of relevance in distinguishing drift and selection. However, if our interest is in quantifying what to expect on average in repeated (though not infinite) cases where selection is not acting alone, then the expected frequency of A’s offspring ought to be used.

Moreover, we argue that, if the probabilities underlying fitness values result purely from abstracting from selectively irrelevant causal features of an organism’s or type’s environment, this also has interesting implications about the relationship between these two measures of fitness. We show that, even when the expected frequency of offspring is the preferred measure to quantify fitness, if certain conditions hold, then [(The expected number of As )÷(The expected number of As+The expected number of Bs)] will be remarkably reliable in predicting evolutionary outcomes.

Beatty J, Finsen S (1987) Rethinking the Propensity Interpretation. In: Ruse, M (ed) What the Philosophy of Biology Is. Kluwer, Boston: 17-30.
Brandon R (1990) Adaptation and Environment. Princeton University Press, Princeton
Gillespie G H (1977) Natural Selection for Variances in Offspring Numbers: a New Evolutionary Principle. American Naturalist 111: 1010-1014.
Sober E (2001) The Two Faces of Fitness. In: Singh R, Paul D, Krimbas C, and Beatty J (Eds) Thinking about Evolution: Historical, Philosophical, and Political Perspectives. Cambridge University Press, Cambridge: 309-321.

Author Information:

Ronny Fernandez    
Philosophy
UMBC

Jessica Pfeifer    
Philosophy
UMBC

 

Powered by OpenConf®
Copyright©2002-2015 Zakon Group LLC