It was not until Wallace's famous bout with malaria that he finally began to extricate himself from the dilemma his earlier logical circumscription had created. But natural selection itself, I now argue, was only step one in the extrication process.
It has long been intimated that Wallace's thinking on the subject of evolution proceeded with uncanny directness in four steps: his adoption of the general evolutionary paradigm around 1845, his recognition of the spatial-temporal implications of speciation (as announced in 1855 in "On the Law..."), his discovery of the natural selection principle as the underlying agent for adaptive change (and thus evolution), and his later rejection of the all-sufficiency of natural selection, perhaps as a result of his adoption of spiritualist beliefs around 1865.
My interpretation of the events, based almost entirely on what Wallace actually committed to print, differs significantly from this scenario. I have already suggested in Chapter One that his eventual adoption of spiritualism was practically preordained as early as 1843 (the time of his writing "The Advantages of Varied Knowledge") or before, and in Chapters Two and Three that his teleological point of view had forced him into the difficult position of coming up with a universal theory that could explain both the peculiarities of individual organisms and their gross properties of distribution and organization without recourse to first-causes lines of reasoning. And there was still the matter of how to fit human beings into the equation.
When Wallace finally saw the connection between the ideas of Malthus1 and the more immediate problem at hand, the implications must have astonished him. It was possible after all to connect a naturalistic argument for utility to the fact of change in particular adaptive traits; indeed, if he wished to be able to generalize at all about the relationship between environment and population change, a greater debt to necessary utility had to be allowed. Admitting character variation, natural selection could be viewed as operating to favor whatever adaptations that could serve to maintain the population through differential persistence. This was a model through which one could understand the process of diversification, as long as one accepted that selection itself (in contradistinction with all individual "accumulations" of character states) was the generalizable law, the necessary result of interaction between life and its surrounding environmental conditions.
As has been pointed out by Browne (1983) and many others,2 Wallace envisioned a selection process operating at the group level, and producing new populations derived from varieties characterized by distinct adaptive suites. Browne notes that Wallace was thinking in terms of statistical generalizations; at least one verification of this surmise appears in the rules for systematic revision (multiple character trait-based analysis, etc.) he was applying at the time.3 If single characters were untrustworthy distinguishers of species from one another taxonomically, neither were they likely to mirror the only relevant population-dividing forces. Suites of peculiarities defined species. In "The Origin of Species and Genera" in 1880 Wallace later discussed this as follows:
Significantly, through Wallace's new model it was still possible to admit that certain individual characteristics might have no adaptive value--as long as their presence was correlated, for reasons yet unknown, with ones that did. This suited Wallace perfectly: he could now explain how new forms came into existence, but could still avoid the assumption that structural continuities over time necessarily implied single causal continuities. In the original printing of "On the Tendency..." Wallace states: "Here, then, we have progression and continued divergence deduced from the general laws which regulate the existence of animals in a state of nature...".5 On the same page he adds: "Variations in unimportant parts might also occur, having no perceptible effect on the life-preserving powers; and the varieties so furnished might run a course parallel with the parent species, either giving rise to further variations or returning to the former type."
The flexibility of the concept extended in other directions that were useful to Wallace as well. Although he could now admit that, perhaps, some adaptations that he thought before were not utilitarian actually were, he still could concede that there were some that were not. These included the same human "higher attributes" he had been struggling to understand for years. That his new understanding seemed incapable of accounting for their existence as well was not that troubling, however: obviously, there were yet aspects of evolutionary final causation he still did not understand.
Otherwise put, it appears to me that Wallace never thought that all levels of functional adaptation could be attributed to the action of natural selection. Those that could were to be contextualized in simple survival terms as responses to the immediate (causal) constraints of environment--but the utility function of the rest had to be referred to associations with causes not yet understood. This conclusion, not natural selection itself, was the real breakthrough in Wallace's thinking. It allowed him a position consistent with his continuing aversion to "first causes" thinking, but conducive to explaining particular one-to-one associations between environment and biological structure.
The relationship that apparently most helped Wallace sort all this out was that existing between man and domesticated animals. Consider the following passages from the original printing of "On the Tendency...":
Wallace's arguments here seem to be directed toward exposing an exception to the rule. The point of interest, however, is that it is not by any means clear which discussion he is posing as the exception, and which the rule. Earlier in the same work he straightforwardly states:
Later in the same essay he notes that his theory explains "the remarkable persistence of unimportant parts such as colour, texture of plumage and hair, form of horns or crests, through a series of species differing considerably in more essential characters."11 His ideas have progressed--ever so subtly--from the non-utilitarian position expressed in the orangutan essay to one in which he views characters varying in being "the more or the less" "essential" (the characters he lists are, of course, among the easiest to modify through selective breeding practices).
The passages I have italicized in the selections above are of especial interest. From the last it is apparent that, in contrast with Darwin, Wallace did not view domestication as a process paralleling natural selection--rather, it was its antithesis. Why? Because no biological "self-regulation" was operating in the case of artificial selection. The latter produced changes "without the animal ever becoming aware of it" or leaving it at any competitive "disadvantage."
In Marchant (1916/1975) is reproduced an exchange of letters between Wallace and Darwin that highlights their difference in perspective on this matter.12 Darwin writes that his "deception" as to the possible significance of single variation was brought about by "simple illustrations, as when man selects." Wallace apparently was already promoting the "antithesis" argument well before the concept of natural selection occurred to him, judging from comments offered by Darwin in a letter of 1 May 1857 responding to an earlier Wallace communication, now lost: "I have acted already in accordance with your advice of keeping domestic varieties, and those appearing in a state of nature, distinct, but I have sometimes doubted the wisdom of this, and therefore I am glad to be backed by your opinion."13
So these examples taken from the realm of domestication were the key: traits that could be passed on, for no apparent reason of utility (that is, with respect to "natural" environment), yet produce no "disadvantage." By inference, biological utility was a valid concept only to the extent that it supported characters whose regulation (or further development) in a population might well be guaranteed only through the intervention of additional, but not yet known, causal agencies. In 1870, in the last essay in Contributions..., "The Limits of Natural Selection as Applied to Man," Wallace in part qualifies those characteristics that can only be maintained through such an additional causal agency as being ones that "transcend time and space."14 Immediately thereafter, he notes parenthetically that "all of [these] were occasionally manifested at such an early period of human history as to be far in advance of the few practical applications which have since grown out of them".
This combination of themes appears, to varying extents, in many of Wallace's writings, beginning well before his natural selection days, and is one of the most obvious indicators of his lifelong commitment to final causes-based thinking. One of his first trains of thought tending in this direction is his treatment of vestigial organs in "On the Law..." in 1855. At one point in this essay he refers to these as "apparent imperfections" and how most authorities recognize them as having no "apparent function."15 He refers to them in fact as "rudimentary organs," an interpretation viewing them as incipient structures rather than remnant ones, as Ghiselin (1969) has pointed out.16 Wallace writes:
This rather "to-the-future-looking" understanding is perhaps one of the main reasons why Darwin ignored the essay initially: it seemed to be invoking design more than process, or at the very least, process on the basis of design.
In 1862 Wallace wrote the following:
Here we see some early indications of the notion that the potential to become civilized may have evolved before the environmental opportunities actualizing such potential. From there it is but a short step to the 1870 essay in Contributions... containing the more specific thoughts mentioned above: "all of [these] were occasionally manifested at such an early period of human history as to be far in advance of the few practical applications which have since grown out of them."
Wallace's most obvious application of the "transcend time and space" line of thinking is to the appearance of mediumistic powers--abilities that he argued that were, like mathematical and artistic talents, not necessary for sake of mere survival, and therefore not explainable through the action of natural selection. Their appearance prior to any use for them bespoke additional causal agencies. In 1888 he wrote:
Similar thoughts by Wallace were reported in a 1904 interview by Harold Begbie.20
There is actually very little distance conceptually between these thoughts and how Wallace felt natural selection operated on purely biological forces. Consider his following treatment of the concept of utility, taken from "Modern Biology and Psychology" in 1891:
As Wallace describes the matter in the preceding quotation, "nascent structures" appear "sporadically," being "at first of no utility," until "under some changed conditions of the environment they become of use." As time went on, into his later years, he increasingly came to believe that such "sporadic appearances" were not so entirely sporadic--that is, that although in any given instance they might not be predictable in their inception, overall there was a progress in structure and organization that was indicative of the action of more remote causes.
Far from a post-1858 phenomenon, this "forward-looking" perspective appears even in Wallace's earliest known writings. In "The Advantages of Varied Knowledge," composed around 1843, he writes:
Similar sympathies are expressed in an even earlier writing, "An Essay, On the Best Method of Conducting the Kington Mechanic's Institution."23
The place of the "additional causal agency" concept in Wallace's approach to natural selection and organic evolution in general is a large enough subject to merit its own monographic study. For the moment, one particular connection can at least be briefly alluded to. This concerns his early reluctance to attribute all aspects of mimetic/protective resemblance to the workings of natural selection, despite the fact that he himself was as important an early contributor to the former theory as his colleague Bates.
The general theory of protective resemblance argues that in many instances a selective advantage is enjoyed by those individual organisms and populations that look like something else, and that the more they do, generally, the more advantage there is. The circumstances of the advantage vary; in some instances it is accrued simply by enhancing camouflage value and thus increasing oversight by enemies, in others by bearing resemblance to another creature that is dangerous or noxious. Henry Walter Bates was the first to work out the logic of mimetic resemblance,24 though Wallace also contributed to the theory's early development in the 1860s. Wallace, on the other hand, was more completely responsible for the elucidation of the various forms of protective coloration as they related to behaviors and concealment circumstances. Despite his basic support of the theory of protective advantage, however, he was initially suspicious of its ability to explain all instances of color and pattern resemblances.
This seemingly rather conservative position (at least for Wallace) was likely a holdover from his pre-natural selection days. During the early 1850s, it will be remembered, he had become convinced that the species divergence process was more relatable to large-scale environmental forces than it was to programs inherent in the adaptive arrays of individual creatures. This rather simplistic brand of spatial/environmental determinism had considerable appeal to him at this point, as it not only explained the geographical/geological record of divergence (as set out in "On the Law..."), but was consistent with his knowledge that areas of similar climate often sponsored both completely different arrays of species populations, and different adaptations to accomplish the same kinds of functions. The natural selection concept made it possible to view these issues through a single lens, but, as his observations of human attributes demonstrated, it did not guarantee that all outcomes, including those perhaps more attributable to the origins of variation than to natural selection itself, were reducible simply to the operation of that principle.
Wallace's suspicions on this score were still considerable as of 1876, when he delivered an address at the annual meeting of the British Association for the Advancement of Science, a portion of which was entitled "On Some Relations of Living Things to their Environment."25 The opening words of this are: "Of all the external characters of animals, the most beautiful, the most varied, and the most generally attractive are the brilliant colours and strange yet often elegant markings with which so many of them are adorned. Yet of all characters this is the most difficult to bring under the laws of utility or of physical connexion."26 Wallace argues in this essay, through the use of many kinds of example, that "local influences" may yet be at work to explain the various characteristics he describes, and his leading advice to naturalists is that greater efforts be made to examine the nature of these. In short, he is still treading a conservative line regarding the extent to which natural selection could be assigned to completely understand "those complex reactions between the vegetable and animal kingdoms, and between the organic world and the inorganic, which have almost certainly played an important part in determining many of the most conspicuous features of living things."27 Indeed, given the fact that the essay does not even discuss the operation of natural selection, one wonders whether his words might represent a nod in a direction allowing for the existence of more remote laws. At the very least, they suggest that he felt that something was missing in the current explanatory power of natural selection vis a vis these "conspicuous features."
A few years later Wallace encountered the writings of Fritz Müller on the latter's newly proposed form of mimicry, involving inedibility, that helped account for many of these apparent exceptions to the explanatory power of protective resemblance. In an 1882 communication to Nature Wallace wrote:
As time went on the logic of natural selection proved capable of providing enduringly logical explanations for more and more such phonemena. But these new revelations apparently never entirely assuaged Wallace's suspicion that more remote causes were still influencing the introduction and modification of characters. Indeed, it would appear that as the years passed, he became more and more convinced that many things were "evolving in spite of themselves," the very notion that suffuses the 1870 essay "The Limits of Natural Selection as Applied to Man" from Contributions... In his late work The World of Life he comes to the remarkable conclusion that the protective/mimetic coloration of many insects arises from selective effects imposed on them by animals of higher organization than themselves.29 In this instance he is invoking a chain of causation not unlike his distinction between natural and "artificial" selection (i.e., domestication and breeding), in that a more organized entity is controlling the characteristics of the selection process within a lesser one.
But again, although he may in fact have become "more and more convinced" of the presence of such remote causes as he aged, these are not merely the product of his old age or even post-1858 thinking. As discussed earlier, there is very clear evidence in his writings from well before 1858 that there were transcendental reasons for believing that the plan of nature extended to beyond our appreciation of its workings.
My general point here is that Wallace probably accepted the various implications of final causes-linked thinking in part because he knew that it would be very difficult in any given instance to prove that an immediate "one cause-one effect" determinism actually existed. Before 1858, he had resisted the evidence of his own field work in this regard for over ten years. After 1858 nothing had really changed in this respect--it still was difficult to prove in any given instance whether a particular adaptation was assignable to a particular program of its environment--and it should not be surprising that he tread very carefully in this direction when it came to trying to defend the notion in detail. For an example of how Wallace handled this aspect of "one cause-one effect" thinking in the crucial "The Origin of Human Races..." essay in 1864, note the following, representing a somewhat elusive response to discussion that arose after its presentation at the Anthropological Society of London meeting:
Here Wallace points out that some correlations, at least, seem to exist between the varying physical nature of man and the varying characteristics of the environment. He does not argue, however, that all characteristics of man must be treated in this fashion.
In the same essay, he uses the loaded term "provident" to describe the development of environmentally useful human characters: "When the power that had hitherto modified the body, transferred its action to the mind, then races would advance and become improved merely by the harsh discipline of a sterile soil and inclement seasons. Under their influence, a hardier, a more provident, and a more social race would be developed."31 The President of the Anthropological Assocation, James Hunt, queried him on this; his reply was: "Why, is it not perfectly clear that if people live in a country where there is a severe winter, in which little or no food is to be had, that they must provide against the scarcity, and that gradually the race would become a provident race? Therefore, I think I am justified in saying that, given two races of the same capacity, and put one in a tropical and the other in a temperate climate, the one in the temperate climate will become the more provident race of the two."32 The use of this somewhat uncommon term, noticed by Hunt, suggests a hidden agenda.
These various connections explain Wallace's continuing interest in the study of character utility and why, as he grew older, he was able to develop an increasingly explicit teleological cosmology while maintaining his strong defense of natural selection. Simply, as of 1858 he could not guage to what extent the immediate physical/biological environment determined utility function--and, in fact, in later years he famously abandoned some of his earlier arguments regarding particular "non-necessary" physical traits, as Kottler (1974) points out.33
I suggest, therefore, that the purpose of "On the Tendency..." has been grossly misconceived. It was not intended in the main as the revelation of a "new theory of evolution" (i.e., natural selection), but as an interim statement clearing the ground for an interpretation of evolutionary change that was not bound a priori by the assumption that all adaptive characters--that is, including the human "higher attributes"--are directly related to the program of the immediate environment. Natural selection thus in effect became the "rule" that distinguished between those characteristics that were so related, and those for which additional explanation was necessary.
Those who are familiar with the ideas of George Romanes on physiological selection will perhaps notice that I am to a degree accepting one of his criticisms of Darwinism--especially the Wallacian brand--in the years following Darwin's death in 1882. In his original essay entitled "Physiological Selection" Romanes objected that natural selection "is not, strictly speaking, a theory of the origin of species: it is a theory of the origin--or rather of the cumulative development--of adaptations, whether these be morphological, physiological, or psychological, and whether they occur in species only, or likewise in genera, families, orders and classes."34 Lesch (1975) writes: "In other words, while the theory of natural selection had succeeded in accounting for the great central fact of adaptation in nature, it had failed to give a complete explanation of the special phenomena associated with species formation."35 I would also agree with Donald Forsdyke36 that Wallace did not entirely understand Romanes's objective, which was not to set forth a mechanism which in any sense replaced natural selection. (It is, after all, one thing to argue that all adaptational arrays are "filtered" through the influence of operation of natural selection, and quite another to argue that all existing structures are created by natural selection: the former suggests in the main a negative feedback process, whereas the latter sounds more like positive feedback.) Indeed, it seems that Wallace might easily have simply accepted physiological selection (assuming some empirical evidence could be offered as proof of its operational existence, that is) as another one of those laws of variation that we did not yet fully understand. Later, however, I will argue that it is this Wallacian notion of natural selection as a model of the "cumulative development of adaptations" that permits us to view the evolutionary process in a manner that more clearly recognizes the difference between process and form.
* * *
That Wallace sent the Ternate essay manuscript to Darwin in early 1858 was entirely predictable: the production of varieties through domestication processes had been one of the main subjects of their earlier exchanges of letters, and here was an opportunity to show not only how such varieties could be produced naturally, but to distinguish the process from its results. One surmises that he felt Darwin would be interested in both parts of his argument. As mentioned earlier, however, Darwin recognized no logical difference between "artificial" and "natural" selection processes. To be sure, the immediately operating influences in each instance were quite different; the common thread for him, however, was that both kinds of selection produced the result of changes in biological form. Wallace, by contrast, still holding the view that changes in form were no more than correlative to process, was coming at the matter from a totally different starting point.
In 1870 Wallace wrote:
In the same essay he later adds:
I see no reason why Wallace might not just as easily have written the preceding words in 1864--or even 1858--as 1870: he merely would have felt a good deal less confident in doing so before he had collected evidence supporting the notion that, in parallel with domesticated animals, man was evolving without "ever becoming aware of it." The essential element of Wallace's cosmology with respect to man--that forces extending beyond his immediate awareness of their operation and impact were influencing his overall development--remained intact between at least as early as 1856 (the year of the orangutan essay) and his 1870 publication of Contributions... It must have been in good part for this reason that Wallace wanted to hear what Darwin and Lyell had to say about his theory before he attempted to publish in 1858: he believed it identified no more than the critical efficient cause of the process named in the paper's title. It did not yet explain, however, why some characters could come into existence and be maintained despite their apparent inutility. Would others--especially Lyell, who Wallace must have known was not an evolutionist--object to this weakness?
Ironically, Wallace was probably afraid that his new theory would be criticized on the grounds that it attempted to explain "too much;" i.e., the dynamics of emergence of all purely biological structures. Some critics of Wallace's theory of natural selection have in fact argued that it did try to explain too much (i.e., is "panselectionist"); others, however, have complained that it explained too little (i.e., irrationally put human change outside the causal realm of the "survival of the fittest"). Wallace was never willing to acknowledge either objection, arguing instead that natural selection was no more nor less than the logically inescapable result of the acting out of individually organized forces within a limited domain. His irritation at this misconception is reflected in the many times he specifically objected to "judgment being passed on a theory of nature by its power to explain all mysteries."39
As it so turned out, the pendulum swung the other way, and with the premature reading and printing of the paper and Darwin's publication of On the Origin of Species the next year, Wallace's initial worry became academic. With no proof of--or even model of the operation of--more remote causes at hand, it would have been futile to press the issue at that point, and Wallace turned his attention to defending ground already gained.
And, in any case, the main directions in which to proceed were apparent enough. Clearly, it was crucial to the more immediate cause of defending evolution and natural selection to develop a better appreciation of the central relationships, and in turn to get others thinking about related subjects and contributing. Just as clearly, no attempts could be made to raise specific objections to Darwin's more open-ended position on evolutionary causality until Wallace had taken the time to consider in detail the possible limits of the explanatory domain of natural selection. Thus, for the next several years his theoretical papers concentrated on connections between biological change and (1) adaptation;40 and (2) geological/geographical distribution.41
It is surely for this reason that Wallace had almost nothing to say about the evolution of man between 1858 and 1864: he had little to add to what he had already said in 1856 regarding the relationship of continuity to utility, and there was no point in speculating further until a solution to the remote causality issue presented itself.
One cannot help but wonder whether Wallace might actually have felt an ironic sense of satisfaction in the fact that natural selection as it is usually conceived is now largely associated with Darwin: indeed, his support of the term "Darwinism" itself may be viewed in a different light. One should not forget the concluding words of Darwinism (1889) in this regard:
In 1852 Wallace had lost his valuable collections to an on-board fire while making his way back to England. With the loss any hopes of thinking out an answer to the origin of species problem through after-the-fact study of his collections (as had been possible for Darwin) had disappeared. Undaunted, he had gamely struggled on and ultimately succeeded in making a great scientific breakthrough. The only problem was that the breakthrough he became recognized for was not quite the one he had made. Should we wonder that he continued to issue his little defensive notes whenever he reprinted or mentioned his epoch-making essay in later years?
Notes, Chapter Four
1. As an aside, the inference by Gordon (1989) that Wallace was not influenced by Malthus in his early years and only picked up on Malthusian notions much later, cannot be seriously entertained. Gordon suggests that Wallace's remembrance of the matter was vague and untrustworthy by 1908 (his discussion of the subject in his presentation to the Linnean Society's fifty years celebration), but Wallace produced much earlier recollections: especially, in an 1869 letter to Adolf Bernhard Meyer, the translator of several of his works into German. See Beddall (1968), p. 313; and a 9 July 1881 letter to Darwin printed in Marchant (1916; 1975 reprint ed.), pp. 260-261.
2. Peter Bowler (pers. commun.) has made the interesting point that the particular subtitles added to the 1870 Contributions to the Theory of Natural Selection version of "On the Tendency..." tend to draw attention away from Wallace's more population-oriented view of natural selection. Whether this was purposeful or not is an open question, since the subtitles may or may not have been supplied by Wallace himself.
3. For example, in S28, published in 1856. Browne (1983), p. 181, notes: "...Wallace did not visualise selection as a process where individual organisms competed with each other directly; he did not possess a Darwinian understanding of events, where living beings vied among themselves... Wallace's selection did not take place at the level of the individual per se, but at the level of varieties and species. He believed evolutionary change was brought about by the replacement of one group by another, by one taxonomic unit taking the place of the next. His varieties (distinct from mere individual variants) were thought to be more suited to the immediate environment than the parent; this was why they increased in number and eventually superseded the original association; when the new species produced further varieties, it too would succumb in its turn. The process was one of increasing adaptation and adjustment of an entire taxon to local environmental factors." See Kottler (1985) for further related discussion.
4. S322, p. 102.
5. S43, p. 59.
6. ibid., p. 59.
7. ibid., p. 60.
8. ibid., p. 60.
9. ibid., p. 61.
10. ibid., p. 54.
11. ibid., p. 62
14. S165, p. 358.
15. S20, p. 195.
17. S20, pp. 195-196.
18. S65, p. 136.
19. S398 (Harbinger of Light reprint), p. 3530.
20. S742, pp. 78-79.
21. S432, pp. 338-339.
22. S1, pp. 203-204.
23. S1a. Though not published until 1845, this essay was apparently composed around 1841, according to Hughes (1989).
25. S257, pp. 101-110.
28. S353, p. 87.
29. This general argument is made especially in Chapters 9 and 15, where it is posed both that the usefulness of color is related both to concealment and to recognition, and that bright coloration itself arises for reasons of its usefulness to higher beings.
30. S93, pp. clxxxiii-clxxxiv.
32. ibid., p. clxxxvi.
35. ibid., p. 487. Mayr (1982), p. 496, relays this idea in slightly different terms, dwelling on the way Wallace invoked the operation of the environment to do its selection (of adaptations): "Wallace's own analysis was in some ways more a study in population ecology than a study in speciation (which he presumably thought he had dealt with adequately in 1855)." I believe Mayr is exactly correct in this assessment.
37. S716, 1st ed., pp. 332-333.
38. ibid., pp. 359-360.
42. S724, 1st ed., pp. 476, 478.