Initial understanding in biological systematics
Segregating understanding into three distinct classes when speaking of biological systematics provides a way to indicate the internested relations that exist in systematics research. In other words, there tends to be a three-tier system of inquiry:descriptive understanding, in the form of observation statements, is a consequence of our desire to attain understanding of sense data by way of the existence of objects with particular properties, and that level of understanding naturally leads to sets of questions answerable by proximate causes, given that we routinely observe individuals at different points in their life history, and with observations among groups of organisms there are additional questions to which a variety of ultimate causes can be applied.
Continued understanding in biological systematics
The relations between the classes of understanding presented above tend to be operationally internested from the perspective that our opportunities to test hypotheses that are (1) descriptive, (2) proximate, and (3) ultimate,respectively, tend to be increasingly more restricted and thus less frequent (Fitzhugh 2010a). Compare, for instance, the ability to (1′) test the descriptive hypothesis that I observe a lizard with three toes (I–III, as opposed to II–IV) in contrast to four toes (I–IV), to (2′) the proximate hypothesis explaining by way of ontogeny the presence of toes I–III in this adult, to (3′) an ultimate (phylogenetic) hypothesis explaining the presence of toes I–III in contrast to I–IV among individuals to which several different species hypotheses apply. Assessing the observation statement would minimally require (1″) bservations of the expected skeletal components that comprise I–III (as opposed to II–IV), while testing the ontogenetic hypothesis would require (2″) the commitment of time and resources to observe causal relations among limbs and toes within individuals over some period of time duringtheir life history. Testing the phylogenetic hypothesis would be the most challenging, as it would require (3″) access to test evidence not only regarding the specific causal events associated with origin and fixation of the threetoe condition in an ancestral population, but also evidence of the cause(s) that led to subsequent splitting(s) of the population(s), colloquially referred to as ‘speciation.’
Assessing our causal understanding in biological systematics, whether descriptive, proximate, or ultimate, is determined by the extent to which hypothesis testing, sensu [3], is accomplished. While the act of inferring an explanatory hypothesis provides initial understanding, in that the hypothesis serves as an answer to at least one specifiable question, it is a hallmark of the sciences that engaging in critically evaluating such hypotheses allows us to not only gauge the current status of understanding but to alter or revise that understanding over time. But as noted earlier, the ability to test biological systematics hypotheses becomes progressively more difficult as one proceeds from descriptive, to proximate, to ultimate explanations. A prominent limiting factor for the testing of higher-level ultimate hypotheses is the span of time between those effects that prompt inferences of hypotheses and the causal events themselves. The greater the span of time from a hypothesized cause(s) and observed effects the more likely will be the eradication of relevant evidence required to test those hypotheses (Cleland 2011a). The consequence of the inherent difficulty with testing ultimate explanations is that two of the most prominent classes of evolutionary hypotheses, specific and phylogenetic (cf. Fig. 1), are almost never legitimately tested, such that
there is no actual enhancement of understanding via hypothesis revision or replacement on the basis of empirical
assessments of causal relations. Rather, once ultimate hypotheses have been inferred, the tendency among
systematists is to direct focus back to the inferences of additional or revised descriptive and proximate hypotheses,
if at all, in a purported attempt to refine ultimate understanding qua testing (see below). For instance, Hennig
(1966: 122, emphasis added) mistakenly considered this maneuver as a process of testing phylogenetic hypotheses
Likewise, the more characters involved in this process the better (Hennig 1966: 132, emphasis added): “For phylogenetic systematics this means that the reliability of its results increases with the number of individual characters that can be fitted into transformation series.” Rindal and Brower (2011: 331; see also Brower 2006, 2010) go so far as to make the claim that character congruence on cladograms is the only means of phylogenetic assessment. Unfortunately, in the nearly 40 years since Hennig’s statements, the emphasis on congruence or acquiring more character data under the guise of testing has grown in prominence, most notably through the peculiar attempts to ally this activity with Karl Popper’s notions of corroboration or falsification (e.g. Wiley 1975; Gaffney 1979; Eldredge & Cracraft 1980; Rieppel 1988; Kluge 1997a, 1997b, 1999, 2001; Siddall & Kluge 1997; Farris et al. 2001; Faith & Trueman 2001; Faith 2004; de Queiroz & Poe 2001; Lee & Camens 2009; Wiens 2009; Faith et al. 2011). As will be outlined in the next section, the result has been the development of a cycle of at best minimizing, and at worst impeding causal understanding as a consequence of conflating the inferences of evolutionary hypotheses with their being tested.
The process of acquiring causal understanding in biological systematics
The previous two sections outlined what is required to proceed from sense data to three general classes of understanding via inferences of explanatory hypotheses. The last section ended with the observation that some authors in biological systematics have developed a protocol that falsely claims to move ultimate understanding forward by actions that are not valid test procedures. The nature of this confusion will be addressed in part in this section. The subsequent section will examine common approaches to evaluating phylogenetic hypotheses, either in the context of testing or stipulating evidential support.
Conclusions
Scientific understanding occurs by way of explanation through the fitting of observations into some broader theoretical framework, not only by offering initial information about possible causes but also through the ability to anticipate and investigate consequences related to those causes as matters of critical evaluation. Understanding is also context dependent in that it is a state of mind contingent on what individuals regard as being sufficient for meeting their standard of understanding. What provides an adequate explanation for one individual might be unsatisfactory to another. As a result, some yardstick by which to judge the adequacy of understanding is required. Surely any modicum of consensus on the adequacy of hypotheses in the sciences should come from the extent to which results from empirical testing are manifested. Therein lay two problems for biological systematics. Ultimate hypotheses, especially specific and phylogenetic, are often devoid of causal details, such that the state of understanding, beyond the initial explanatory notions presented under the rubric of ‘taxa’ or cladograms are neither pursued nor enhanced. Even if these hypotheses are filled out to the point that valid test predictions can be stipulated, it is likely that actual testing will be impractical in nearly all instances, as noted earlier. Instead, critical assessments of hypotheses are stalled, such that there is the tendency to orient back to original character observations, such as interesting correlations among features, or pursuing investigations of the finer structural components of characters (descriptive) or their ontogenetic development (proximate). In other words, the general reaction is to move backward in an epistemic sense to consider the enhancement of descriptive and proximate understanding, rather than actually pursuing ultimate understanding in terms of critical hypothesis assessment. Taken at face value, there is nothing wrong with such a maneuver, but we do need to be cognizant that enhancing descriptive or proximate understanding does nothing to promote continued ultimate understanding as conceived as testing in the sciences. Several of the classes of systematics hypotheses fail to provide substantive growth in causal understanding for the fact that our tendency is to maneuver away from testing those hypotheses.