Two Evolutionary Theories--A Discussion Author(s): Walter J. Bock Source: The British Journal for the Philosophy of Science, Vol. 14, No. 54, (Aug., 1963), pp. 140-146 Published by: Oxford University Press on behalf of The British Society for the Philosophy of Science Stable URL: http://www.jstor.org/stable/685431 Accessed: 08/05/2008 15:45 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=oup. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.
JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We enable the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact
[email protected].
http://www.jstor.org
DISCUSSIONS Two
THEORIES-A DISCUSSION EVOLUTIONARY
IN the pages of thisJournal(1959, 9, II0-I27, I85-I93) under the title 'Two Evolutionary Theories', Marjorie Grene analysed and compared the mechanisms of evolutionary change advocated by Simpson in his MajorFeaturesof Evolution(1954) with those proposed by Schindewolf in his Grundfragender Palaontologie(I950). Tiis choice of works is an excellent one as these theories are directly opposed to one another in both their foundation and general philosophical significance. As a result of her study, Grene concluded that Schindewolf's theory is more adequate than Simpson'sand more in agreementwith her philosophicalideas-that a historical form of Kant's transcendentalmethod must be applied in studies of evolutionary theory. We find ourselves, as students of' macroevolution', in sharpdisagreement with Grene's argument and conclusions. Lack of space prevents us from delving into all of the interestingproblems arisingfrom her study; we shall touch only upon the most important ones. The interested reader is referred to Schindewolf's and Simpson's works (and the literature cited by them) for scientific details and the evidence supporting our comments; we feel that an elaborate discussion of these data would be out of place in our note. We choose to substitutethe term ' Synthetic theory of evolution' for Grene's' Neodarwinism ' as the former is the more acceptedterm for the school of evolution representedby Simpson; ' Neodarwinism ' has been used for several differentperiods of evolutionary thinking and while most evolutionistsare Darwinists,there is considerabledisagreementon the exact meaning of the term Neodarwinism. Evolutionand phylogenyare two of the most used as well as the most misused terms in evolutionary biology. Their exact definitions are rarely given as their exact use is frequently impossible to ascertainwhen reading an evolutionary treatise. Many students believe that evolution and phylogeny are absolute synonyms, when in fact they representtwo sharply distinct ideas.1 Unfortunately, a precise clarificationof these terms and of the various aspectswithin evolutionary biology is lacking. Thus it is not surprisingthat Grene freely interchangedevolution and phylogeny, and claimed that evolution is a historicalscience. It is most unfortunate that Grene was misled into this false position by the recent evolutionary literature becausethe main factor preventing her from discovering the real differencebetween the rival theories is her confusion of evolution and phylogeny. Any study of a historical subject, be it organic evolution, geology, western civilisation or whatever else, has three major aspectswhich are: (a) The description of the sequencesof events and changes( = historicalconfigurationalsequences)which have occurred during time, these events are only described or listed without 1 Consultingstandarddictionariesis of no help for they follow the currentpractice of evolutionistsand report the usual confusion between these terms. Indeed, analysis of the definitionsgiven for evolutionand for phylogenyin the secondandthirdeditions of Webster's New InternationalDictionary reveals that these terms are regardedas close synonyms of each other by the compilersof this work. 140
TWO
EVOLUTIONARY
THEORIES-A
DISCUSSION
comment-the listing usually being in a chronological order; (b) The study of the with other mechanismsand processes,including causes,consequences,interrelationships influences, etc. (=immanent propertiesand processes),by which the subject matter changes; and (c) The overall explanation of the events and changes which have occurred using the known mechanisms of change. The first aspect concerns itself with a pure descriptionof all the events that have happened and is a historicalstudy. The second deals with how modificationscan take place, how one set of events can cause a certain change or what one would expect if certain events had occurred. Although this branch deals with historical subjects,it is not a historicalstudyin itself. The fact that the mechanismsand processesmay operate over a period of time does not automaticallymake their study a historicalone. Nor does their deduction from historicalsequencesmake their study historical. The last branch,like the first, deals with the actual occurrences that had taken place and is also a historical study. Indeed, it is the ideal historical study for it includes all possible aspects, being a combination of the first two parts. We have listed all three aspects as separate subdivisionsof historicalstudy although other workerswould preferto place ' a' and ' b' as subheadingsunder 'c' which they would regard as the complete historical study. To argue strongly for one arrangementor anotherwould be quibbling. We shall follow the tripartitesystem becausewe feel that ' c' is not a simple combination of 'a' and 'b' and because the tripartitesystem permits a better clarificationof evolutionary biology which is our primary concern. The entire field of evolutionary studies should be classifiedinto these subdivisions and names assigned to them. We would like to emphasisethat this division is for purposes of clarification only and does not represent actual and separatebranches of inquiry. The overall field can be labelled as 'Evolutionary biology' and is divided into: (a) The descriptionof all lineages of animalsand plantsfrom the origin of life to the present, the events being placed on a proper time scale-this may be called the study of phylogeny; (b) The study of the mechanisms of evolutionary modifications in organisms, be they large or small changes-this may be called the study of evolutionary principles,or the study of evolution; and (c) An explanation of the changes which have occurred in all lineages of animals and plants using the known evolutionary principles-an accepted term for this aspect is lacking, but it may be called evolutionary (or explanatory)phylogeny or the study of the evolution of animal and plant groups. Some workers would distinguish only between 'a' and ' b 'which together would comprise' c' or evolutionarybiology. Again we feel that this leads more easily to confusion. If one is concerned with the study of evolutionary theory or a principle of evolution, then one is dealing with the second subdivision of evolutionary biology which is not a historicalscience. Genetics and population biology fall into this category. When one speaks about the phylogeny of ananimalgroup, or of a structurein an animalgroup,then thefirstbranchis implied, while the evolution of ananimalgroup or structurefallsunder the third heading. This distinctiondeservesfurtherclarificationas it is frequentlyconfused. A puredescription of animal lineages-which group follows which or which group gives rise to whichcan be given without offeringany explanationfor the observedchanges. This type of studyis commonly done in paleontologicalwork. Itisphylogeny andsuchwork should never be entitled 'The evolution of. . .' as is frequently done. When, however, the sequencesof groups are given with explanationsfor the observed modifications, 141
W. J. BOCK
AND
G. VON
WAHLERT
why new groups arose and why old ones died out, then one is looking for causes and must use the known principles of evolutionary change. Here one is really concerned with the evolution of animal groups and these studies can rightfully be called 'The evolution of. . .'. Both studies are concernedwith phylogenies and are truly historical sciences. These remarkscan be summed up by the following distinction: While the study of phytogenyis a historicalscience,the studyof evolution (= evolutionarytheory)is clearlynot historical. From these comments, it should be quite clear that for the most part, Grene was interestedin theories of evolution, not in phylogeny; the title of her paper is 'Two Evolutionary Theories'. The pertinent differences between Simpson and Schindewolf are not in phylogeny-they both use the same knowledge of animal lineages; they differ in the evolutionary theories deduced from these facts. The facts used by Simpson and Schindewolf are historicalones, but the processesdeduced from them are not. The processesare not historical,in the sensethat they themselves do not change with time, but only their effectsas seen in the configurations. Thus, Grene is wrong in saying that the study of evolutionary theory is a historicalscience (p. 187). Her methodological criteria under philosophical reflections,namely that the historical method is the necessaryone in studies of evolutionary theories, must be rejected. It is still correct that the historicalmethod is the basic one in all phylogenetical studies, both descriptiveand explanatory. one excludes the other. This Morphologyand the synthetictheoryare incompatible, is one of the strangestideas advanced by Grene (pp. I15-II6) and one that must be rejected as completely false. It may be true that Schindewolf's morphology is incompatible with the synthetic theory, but he representsa very small segment of present morphological thought. Grene shows confusion, for example, in her assumption that morphology must be typological. She consequently fails to see that it is typology, not morphology, that is inconsistentwith the synthetic theory. The basic ideas used by most morphologists today are in complete agreement with the synthetic theory of evolution, if not based completely upon it. The synthetic theory is not founded upon the determination of ignoring structure, it is deeply concerned with examining and understandingstructure as structure as well as its genetical and development bases. It is, thus our belief that by accepting Schindewolf's position as typically representativefor all morphologists, Grene was badly deceived about the true relationshipbetween morphology and the synthetic theory and was unable to understandSimpson'sinquiriesinto the genetical basesof structure and of macroevolutionarychanges. betweenthe two rival theoriescouldbe eliminatedif the different Muchof the difference which Grene claims is possible and suggestsbe done sets of termsusedwere translated (p. II7). She notes that Schindewolf and Simpson would object strongly to this suggestion, to which we would add, quite rightly so because they know the connotations of the terms as well as their meanings. Grene believes that many terms such as ' orthogenesis' could be interchangedfor 'essentially rectilinearevolution' or ' generalimprovement' for 'general adaptation' without loss of meaning. This cannot be done because the implications of these terms go further than their definitions. Schindewolf, for example, implies much more than mere description when he uses such terms as ' type', 'general improvement', ' bauplan' and 'orthogenesis'; these additional implications are objectionableto students of the synthetic 142
TWO
EVOLUTIONARY
THEORIES-A
DISCUSSION
theory. And Schindewolf would object to the connotations associatedwith the terms used in synthetic theory. Both know that there is more to their terms than the explicit meaning. Perhaps these connotations should be clearly stated by the author when he defines terms; however, this is not done, and really cannot be done without making the scientific literature completely boring. A knowledge of the hidden meanings is expected of those reading the scientific literature. Because these hidden meanings are among the most subtle parts in the evolutionary literature, extensive reading is required before they can be easily recognised and separated. Simpson'sand Schindewolf's terms are strikingly differentin their connotations,and we disagree with Grene that the verbal differencesbetween their theories would disappearif each term was interchangedfor the 'corresponding 'one. The verbal differencesare not the result of a simple use of differentwords, but are the result of clearly separableideas which are expressed by the use of different terms-words which cannot be translated. Analysis of ontogeneticchangeand correlationbetweenontogeneticand phylogenetic change form the basis for much of Schindewolf's theories.
Grene (pp. II9,
122-I23)
acceptsthese ideas and summarisesthem into the concept that phylogenetic relationships can be shown with 'reliance on analogies drawn from embryology', to which we object strongly. Phylogeny and ontogeny are both time oriented changes, but they are basically different processes. Ontogeny is the change in an individual from its formation until its death, while phylogeny is the change of organisms through a succession of generations. For sexually reproducing species, phylogeny must be thought of as a population phenomenon. Phylogeny may be considered to be a succession of ontogenies, and alteration in ontogeny certainly form the basisfor phylogenetic changes. But it is not possible to ascertainwhat has happened during past phylogeny by examining the sequence of events in an ontogeny, nor is it possibleto claim that ontogenetic changescan tell us anything about phylogenetic changes-that these processesare subdivisions of a major overall process and hence that modification in one (ontogeny) means automatic and equivalent modification in the other (phylogeny). We feel that the burden of proof restslargely with those who still claim such a relationship. The importance placed by Schindewolf on early ontogenetic change is based on the idea that the earlier a change occurs in ontogeny, the greaterwill be the resulting modification in the adult structure-and hence, for Schindewolf, the greaterwill be the taxonomic change in the type. This argument is grounded upon the same basic thinking that has led to the formulation of the ' Biogenetical law' (Ontogeny recapitulatesphylogeny) by Haeckel, in spite of the fact that Schindewolf argues strongly againstthis law. Both conceptsarebuilt on the belief that featuresappearing earliest in the phylogeny of a group appearfirst in the ontogeny, followed by the featuresemerging later and later in the phylogeny. That is, the featurescharacteristic of the phylum appear first in embryological development, followed by those of the class, order, family, genus and last of all the features of the species appear. What appearsfirst in ontogeny must also have appearedfirst in phylogeny. The argumentsput forth by de Beer and othersagainstthe Biogenetic Law can be applied equally well against the ideas of Schindewolf. There are other areasin which Grene displays a lack of understandingfor the facts of modern biology by adhering too closely to Schindewolf's book, but space T43
W. J. BOCK
AND
G. VON
WAHLERT
does not permit us to delve into these points. A major problem restswith Schindewolf's and Grene'sfailureto realisethat evolutionary change in sexually reproducing animals must occur within an interbreeding population. Individuals are always members of a population and cannot singly form the basis for any evolutionary changes. Evolution is the modification of a group of interbreeding organisms, not of independent individuals. Here lies the main error in Schindewolf's theory. Grene has not appreciatedthis failure and hence falls into the same trap that caught Schindewolf. The concept of Typostrophism advanced by Schindewolf illustrates his strict thinking of evolutionary change in terms of modification in individuals with a complete disregardfor populations. The basisfor the theory of typostrophismlies in Schindewolf's concept of' type' and' bauplan'. Type is defined by Schindewolf(p. 240) as the characteristicfeatures of a taxonomic group. Bauplan is almost synonymous with type, perhaps it is best defined as the arrangementof the characteristicfeatures in the animal. This definition is purely descriptiveand is not differentfrom the concept of the characteristics of a group or its morphological definition used in the synthetic theory. But, Schindewolfputs more into his type concept which is not mentionedin his definitions. These additionalconnotations are never defined or clarifiedin spite of their forming the most significantpart of the type concept. Let us first call attention to the fact that Schindewolf defines the type on the basis of the taxonomic group-that is, the type is the sum of the diagnosticfeaturesof a taxonomic group. Most workers agree that the limits of taxonomic groups are establishedrather artificially. This is not to say that the groups are not real, but that their exact limits and which forms are included in a certaingroup, as well as the taxonomic rankingof this group, depend upon the opinion of the individual worker. Hence what one worker would call a family, another may regard as a genus, or what one worker would call a genus, another may divide into two genera. Thus according to Schindewolf, the type or the bauplan would vary according to the opinion of the worker defining the taxonomic group. And since taxonomic groups are erectedaccordingto the known variation of morphological characters,care must be taken when using the limits of these groups to establish the possible limits of variation of types-one can very easily fall into the trap of circular reasoning. Schindewolf goes past his type definition when discussing the type border or limit (Typengrenze, pp. 249, 305-306), and type-defining characters(typenbestimmendenMerkmale, p. 249), or in using specific characters,generic characters,etc. (p. 268), yet never offering one word of definition for these additional parts of his type concept. In the absence of such clarificationswe shall try to deduce what he means by these terms. The type border is not simply the observed limit of variation in a taxonomic group, but is an absolute,hardand fastboundary-the real limit of the type. When the boundaryof the type is passed,which is accomplishedonly by specialevolutionary steps, a new type is reached. The type limit cannot be stretchedor enlarged, it is fixed. The main or defining characterof the type is not simply a useful feature for recognising that particular taxonomic group-a key character-but it is the basic feature of the type; the feature which appearedwhen the new type evolved by the single step jump from the old type-the characterby which the new type is recognised. Species specific characters,generic charactersand so forth are obvious-they are the main features of the species, genus, and so forth. According I44
TWO
EVOLUTIONARY
THEORIES-A
DISCUSSION
to Schindewolf, these are charactersof increasing importancewhich one can recognise, a priori, as being the features of a new species,genus and so on, but nowhere does he give any hint how one recognises these charactersand why they indicate taxonomic groups of different ranks. Most workers agree that there is no such thing as a ' speciescharacter' or ' generic character', but that the group is established first and that the group gives the character,not the charactergiving the group. According to Schindewolf's concept of the type, animals intermediatebetween types cannot exist, they must belong to one type or another. And the change from one type to another must be a single step jump, otherwise there would be intermediate forms and these cannot exist. In this initial step-the jump from one type to another-the main feature of the new type must appear,otherwise it would not be possible to recognise the new type. Hence the main charactersof a new order appearin this first step and in the first individual of the new type if the change happened to be one of ordinal value. Herein lies the problem. An individual may cross the type boundary because of some early change in its ontogeny resulting from a large mutation. Even ignoring the need for explaining the large mutation, and how this radicalchange in one or a few characterscan function harmoniously with the other featuresof the organism and in the possible environment of the new animal, one could ask with which individualswould this single member of the new type interbreedif it is a sexually reproducing organism? The chances of a second individual of the opposite sex making the 'type jump ' at the same place and same time is infinitesimallysmall. Thus, in the large majorityof cases,the single individual of the new type canonly interbreedwith members of the old type, leaving the question of what the offspringwould look like andwhether the mutationleading to the ' type jump ' would show Mendeliansegregationwhen the new type bred with membersof the old type. We also wonder how the appearanceof a new type would allow the animalto be momentarilywithdrawnfrom the confininginfluenceof selection(Grene, p. I23), selectionbeing always the result of the interactionbetween the organismand the environment. No selectionmeansno environmentwhich violatesthe basicconcept that organismsmust always be consideredin relationshipto their environment at all stages of their life cycle. These shortcomings of Schindewolf's typostrophism emphasisehis neglect to regard animals as individuals grouped into interbreeding populationsliving in an environment which are basic prerequisitesfor any theory of evolution. Schindewolf's evolutionary theories present little problem for students of evolution; they have the background to judge them and accept what they believe to be correct. They do present a far greater danger to students of the philosophy of science. Grene (pp. I9I-I93) concludes that the ideas of Schindewolf are simpler, include an ordering principle and a concept of novelty or originality which are more in agreementwith her philosophical beliefs. Such a conclusion is not really surprising,but it may also be meaningless. Schindewolf's evolutionary theories are derived directly from the earlier ideas of the German 'Naturphilosophie' and hence have a very definitephilosophicalbasis. Naturphilosophiehas been strongly influencedby, if not completelyderivedfrom Kant'stranscendentalism. Thereforeto argue,as does Grene,thatSchindewolf'sevolutionarytheoriessupportherphilosophical conceptswould be circularreasoningif both theoriesstemfrom Kantianism. If a philosopher wishes to use a set of scientific ideas as independent support for certain 145
W. J. BOCK
AND
G. VON
WAHLERT
philosophicalconcepts,he mustfirstestablishbeyond any doubtthatthesescientificideas arenot basedor strongly influencedby these philosophies. Grenehas not ascertained, or even inquired into the historical background of Schindewolf's evolutionary theory when making her claim that it offers better support for her philosophical beliefs. Even if her analysis of the rival evolutionary theories was correct, her argument for favouring Schindewolf appearsto be completely circular and hence is incapableof solving the major problem set forth in the beginning of her study. WALTER
J.
BOCK
GERD VON WAHLERT
ON EVOLUTIONARY THEORIES
M. GRENE1 has presented a comparison between the evolutionary theories outlined by Schindewolf2 and Simpson.3 She concludes that there is little biological difference between them and that Schindewolf's is preferablephilosophically. I wish to take issue with both these points, and add certaingeneral comments. Schindewolf's approachis to a considerableextent that of an idealistin the tradition of Plato, Goethe, and Owen, while that of Simpson is more purely materialistic,an approachbeautifully analysedin its relation to evolution by Beckner.4 It is at this point that the first fundamentalbiological differenceappearsbetween the two theories. According to Schindewolf, a higher taxon (a classified group of organisms) actually exists when its first speciesappears. According to Simpson, a higher taxon does not exist (except retrospectively)until it has diverged or diversifiedsufficiently for it to be classifiedas separatefrom its immediatelyancestraltaxon of the same rank. The former view implies that higher taxa exist objectively, with sharpboundaries; the latter view holds that higher taxa have no exact naturalboundariesin time and that more than one valid classificationis often possible on the basis of the same biological information even if this informationis complete.5 Secondly, although related to the distinctionjust made, the mechanismsadvocated for the origins of higher (supraspecific)taxa, and species as well, are utterly different. Schindewolf proposes that each taxon of whatever rank originates full' mutation blown (miteilnemSchlage)from a precedingone by a single' typostrophische and that the resulting individual or individuals of a large effect (Grossmutation), (' hopeful monsters' in the phrase of Goldschmidt)6 are at their appearanceimmediately reproductively isolated from the parentalpopulation. Simpson, on the other hand, believes that except for the relativelyuncommon situationsof polyploidy and neoteny (the former a very differentmechanism from Schindewolf's, the latter only partly compatible with Schindewolf's and leading rapidly to a regressioninto 1 This
Journal, 1958, 9, IIo and I85 Grundfragender Paldontologie,Stuttgart, 1950 3 The Major Featuresof Evolution, New York, I953 4 The Biological Way of Thought,New York, i959 5 See Simpson, Principles of Animal Taxonomy, New York, 1961. All references to this book are more elaborate statements of positions taken by Simpson in 1953 or earlier. 6 The MaterialBasis of Evolution, New Haven, 1940 2
146
