Methodologies and Methods
The epistemic component of a rationally developed cognitive situation includes methodology as one of its most important parts. Its main task is to substantiate and develop methods as specific research tools built into the respective argumentation schemes. Since scientific research is impossible without methods and the development of methods is impossible without methodology, each research program, in a certain sense, can be thought of as a methodological program [Lakatos 1978]. The latter is a specifically organized general regulator of cognitive activity; it is a kind of heuristic that includes general methodological principles, particular methodologies, and finally concrete methods implementing them. It can be divided into two main components: positive heuristic indicates what is needed to be done while negative heuristic indicates what not to do in research so that its results can be considered scientifically consistent cognitive models of natural phenomena.
There is a lot of methodological programs: they are developed by science as a whole, by its different disciplines, and within them by particular research programs. In systematics, the research is generally based on the classification methodology [Rozova 1986] associated with the substantiation and development of technical means allowing for a representation of the structure of taxonomic reality by means of classification systems. The conceptual history of systematics began and proceeded with the development of this methodology and the “natural method” implementing it. Therefore, the main stages of this history were largely marked by the development of particular versions of classification methodology.
The structure of scientific method is composed of three main components, viz. conceptual, logical, and operational [Lukashevich 1991]. Its conceptual component corresponds to the ontic component of a cognitive situation; it is defined by the content of the respective conceptual space; in systematics, it depends on the typological, phylogenetic, phenetic, or other background of the respective taxonomic theories. Its logical component is a part of the epistemic component of a cognitive situation, according to which the content of a method is determined, first of all, by the major cognitive regulators (deductive v.y. inductive, reductionist vs. holistic, etc.). In particular, the procedures of taxonomic research can be based on binary or multi-state logic; they should be substantiated taking into account the principles of selectivity, sampling, etc. Its operational component is “technical”: it contains specific procedures of extracting and processing information about the studied object; its instrumental component largely depends on the technical capabilities of any research.
Scientific Status of Methodologies and Methods
In a rationally organized cognitive activity, the most general regulator of both methodologies and methods is the principle of scientificity, according to which the scientific character of knowledge is determined largely by the scientific character of the methods by which it is inferred. In this regard, the main question becomes what exactly makes a method scientific; obviously, possible answers depend on the basic onto-epistemologies. Therefore, in systematics, understanding of what the “natural method” should be changed significantly in the course of its conceptual history following changes in its onto-epistemic foundations.
In determining the scientific character of the methods under the provisions of non-classical science, the starting point is the principle of onto-epistemic correspondence, from which the principle of methodical correspondence is explicated [Pavlinov 2018]. This means that the general parameters of research methods must comply with the requirements of the onto-epistemic model recognized as scientifically sound within a particular cognitive situation. To this, the principle of method effectiveness is added: a method should allow the research task to be solved effectively within the framework of the respective model according to certain criteria of effectiveness. The more a method meets these two conditions, the more reasonably (other things being equal) it can be thought of as scientific—certainly, not in general but within the framework of the given cognitive situation.
The ontic correspondence of the method can be justified in two ways, direct and inverse; a good example is provided by the methods of phylogenetic reconstructions [Pavlinov 1990, 2005]. The main requirement for them is the correspondence to the postulated properties of the phylogeny. The direct justification of the method means the derivation of the respective classification algorithm from an adopted ontic model through the latter’s operationalization. The cladistic analysis was developed in this way by inferring its working principles (synapomorphy etc.) from certain postulated properties of the phylogeny. The reverse justification implies selection of suitable methods from a variety of existing ones based on an assessment of their correspondence to the respective ontic model. For instance, among hierarchical clustering algorithms, the most suitable for phylogenetics are those that allow, in contrast to phenetic ones, the “arrow of time” to be brought into a tree-like scheme so that it represents phylogeny.
The epistemic correspondence of the method is largely determined by its logical component. So, since scientific knowledge is probabilistic, the methods developed on the basis of probabilistic or fuzzy logic are preferable to the “exact” ones operating with binary logic. Proponents of the mathematization of systematics assess the scientific character of quantitative methods with reference to their mathematical validity [Jardine and Sibson 1971; Dunn and Everitt 1982]. However, emphasis on such a formal assessment of the numerical methods usually decreases the significance of their ontic correspondence and eventually leads to a reduction of the biological content of taxonomic research.
The effectiveness of the method means its ability to solve a specific research task with certain required precision within a reasonable time. For example, in phylogenetic systematics, the effectiveness of its methods depends on their ability to reconstruct the phylogenetic pattern. Thus, this important characteristic is context-dependent: its assessment depends on how the respective Umwelt is defined.
A special emphasis on analytical methods in some methodological programs gives rise to the already-mentioned problem of instrumentalism [Rieppel 2007; Pavlinov 2018]. This means a certain “closure” of cognitive activity to the method as such: the latter dictates how an Umwelt should be analyzed, so the properties of the former indirectly shape the properties of the latter; this is an unpleasant side effect of the above principle of methodical correspondence. For instance, an application of the hierarchical classification algorithm necessarily provides a hierarchically arranged classification, even if the real diversity pattern is in fact non-hierarchical; this means that the respective classification appears to be inadequate as a cognitive model of taxonomic reality.
In considering the methodological problems of systematics, the principle of method limitation is of particular importance: it asserts the impossibility of developing a universal method that is capable of solving all the conceivable tasks associated with the analysis of the taxonomic reality in all varieties of its manifestations. So, with the help of each particular method, only a certain (not any) manifestation of taxonomic reality can be effectively investigated. The causes of the method limitations are of three kinds—conceptual, logical, and operational—which correspond to the three main components of the method structure considered above. This circumstance encourages methodologists to develop new means for solving the particular research tasks as the latter diversify; this results in a growing variety of both the methods themselves, each with its own advantages and disadvantages, and the classifications obtained on their bases. This principle is one of many manifestations of scientific pluralism which also involves methodological programs.