Lloyd Morganís Canon: A History of Misrepresentation
© Roger K. Thomas 2001
Roger K. Thomas, Ph.D.
The University of Georgia
Links are introduced when appropriate to References and four Tables which appear at the end of the article. The links are subsequently interspersed thereafter for those who like to check back and forth as they read. If you link to a Table or to the References, please use the "Back" button on your browser to return you to the place in the text from which you linked. Be advised that while printouts of this article for personal or instructional use are welcomed, the columns in Tables 1 and 4 may be misaligned.
In no case may we interpret an action as the outcome of the exercise of a higher psychical faculty, if it can be interpreted as the outcome of the exercise of one which stands lower in the psychological scale. (Morgan, 1894, p.53)
"Perhaps, the most quoted statement in the history of comparative psychology is Lloyd Morganís canon." (Dewsbury, 1984, p. 187). "To this it can be added that perhaps the most misrepresented statement in the history of comparative psychology is Lloyd Morganís canon." (Thomas, 1998, p. 156)
The most frequently cited source for Morganís canon has been Conwy Lloyd Morganís (1852-1936) An Introduction to Comparative Psychology (1894). However, the canon appears to have been published first in Dixonís synopsis of Morganís paper, "The Limits of Animal Intelligence," presented at the 1892 meeting of the International Congress of Experimental Psychology (Dixon,1892). Dixon referred to it as a "rule" and noted that any quotations were from a written prťcis distributed by Morgan at the meeting.
The misrepresentation of Morganís canon by associating it with the law of parsimony, for example, began at least as early as Stanleyís (1896) review of An Introduction to Comparative Psychology and has continued, at least, through Dewsbury (2000), Leahey (2000, 2001), and Schultz and Schultz (2000). Associating Morganís canon with the law of parsimony has been, perhaps, the most frequent and persistent form of misrepresentation.
Although Adams (1928) was among those who erroneously believed that Morgan intended that his principle be a canon of parsimony, Adams was also among the first to show that they are "...not related...and may on occasion work to exactly opposite effect." (p. 242) Several writers (e.g., Burghardt, 1985; Costall, 1993, 1998; Costall, Clark, & Wozniak, 1997; Gray, 1963a, Miller, 1962; Nagge, 1932; Newbury, 1954; Singer, 1991, Thomas, 1998, and Wozniak, 1993) have tried to correct the association of Morganís canon with the law of parsimony as well as other forms of misinterpretation that will be identified below. Table 1 presents adjacent chronologies of (1) sources that have misrepresented Morganís canon and (2) sources that have tried to correct the misrepresentation of Morganís canon. Illustrative quotations from each of the sources cited in Table 1 may be seen later in this article in Table 2 where quotations associated with misinterpretations and misrepresentations are shown, and in Table 3, where quotations associated with efforts to correct the misinterpretations are shown.
The enduring misrepresentation of Morganís canon provoked Costall (1998) to write, "The extent to which the intentions of Morganís canon has been misinterpreted is astonishing." (p. 18) and Wozniak (1993) to suggest, "It would be an interesting study in itself to trace the progressive distortion of Morganís views and in particular the attribution to Morgan of the principle of parsimony." (p. x) The present work likely will not be the study Wozniak envisioned, but it will contribute to better understanding of the history of misrepresentation of Morganís canon.
How Has Morganís Canon Been Misinterpreted?
Morganís canon has typically been misrepresented in one or more of the five following ways: (1) that it provided a simplicity criterion, that is, that one should prefer the simplest interpretation of animals' psychological processes. In this regard, the canon has often been misrepresented (2) as psychologyís version of Ockhamís or "Occamís razor" and/or (3) the "law of parsimony." Additionally, Morgan's canon has been misrepresented as being (4) anti-anthropomorphic and/or (5) anti-anecdotal. Usually associated with 4 and 5, it is often said or implied that is was written in reaction to the works of George John Romanes (1848-1894) especially Romanes (1882; but see also 1883, 1887). The law of parsimony and anti-anthropomorphism misrepresentations have been the more frequent ones, and the emphasis in the present paper will be on them. However, some consideration will be given to the others.
What Morgan Intended by the Canon
It is a minor point, but in the present context, it should be noted that Morgan did not call his statement a canon at the time that he first stated it, rather he referred to it as a "basal principle" and, as noted earlier, Dixon (1892) referred to it as a "rule." Later, Morgan (1900, p. 270) did refer to it as a "canon of interpretation." Whether he was the first to use "canon" in this context has yet to be determined (by me). Since it is best known as Morganís canon, rather than Morganís principle, unless referring to Morganís references to it as a "principle," it will be identified hereafter as Morganís canon.
In An Introduction to Comparative Psychology, Morgan (1894) addressed several principles that should guide oneís study of comparative psychology. Near the end of chapter III, "Other Minds Than Ours," he wrote:
There is one basal principle, however, the brief exposition of which may fitly bring to a close this chapter. It maybe thus stated:--In no case may we interpret an action as the outcome of the exercise of a higher psychical faculty, if it can be interpreted as the outcome of the exercise of one which stands lower in the psychological scale. (Morgan, 1894, p. 53)
The remainder of his chapter III (viz., pp. 53-59) was devoted to explaining this basal principle, i.e., Morganís canon. Although he retained the exact wording in the italicized version of the canon on page 53 in the 1903 Revised Edition, Morgan added a paragraph on page 59 of the latter in which he offered a substitute version for those who found the phrase "psychical faculty" to be "too reminiscent of a faculty psychology." The revised version was interesting also for the additional modification that may be seen in the last five words. The modification indicates that Morgan was likely already concerned about misinterpretations of the canon. The modified canon was:
In no case is an animal activity to be interpreted in terms of higher psychological processes, if it can be fairly interpreted in terms of processes which stand lower in the scale of psychological evolution and development. (Morgan, 1903, p. 59)
As Morgan made clear in the revised version of the canon quoted above, the higher and lower psychological processes referred to a psychological scale presumed to result from psychological evolution and development. A difficulty long associated with Morganís canon is that he did not provide a psychological scale per se, although one may be inferred from his chapter headings, their order of appearance within the book, and his discussion of the topics of those chapter headings. See Burghardt (1985) for additional discussion of the addendum to the 1903 revision of the canon as well as other useful information not provided here.
Morgan on simplicity as a criterion. As already noted, the most persistent misrepresentation of Morganís canon has been that lower and higher processes are related to some kind of simplicity criterion, typically, such as that embodied in Ockhamís razor or the law of parsimony. Burghardt (1985), as have others who provided critiques of Morganís canon, emphasized that Morgan dismissed the utility of a simplicity criterion. In view of this persistent misinterpretation, it is especially noteworthy that Morgan thought that an objection to his principle was that it might Ďshut outí the "simplest explanation," not that it might should foster a preference for the simplest explanation!
Immediately following his presentation of the canon on page 53, Morgan (1894) began to anticipate and address some of the objections that could be raised against it. On page 54 Morgan wrote a lengthy paragraph that, for present purposes, is worth quoting in full.
A second objection is, that by adopting the principle in question, we may be shutting our eyes to the simplest explanation of the phenomena. Is it not simpler to explain the higher activities of animals as the direct outcome of reason or intellectual thought, than to explain them as the complex results of mere intelligence or practical sense experience? Undoubtedly, it may in many cases seem simpler. It is the apparent simplicity of the explanation that leads many people to naively adopt it. But surely the simplicity of an explanation is no necessary criterion of its truth. The explanation of the genesis of the organic world by direct creative fiat, is far simpler than the explanation of the genesis through the indirect method of evolution. The explanation of instinct and early phases of intelligence as due to inherited habit, individually acquired, is undoubtedly simpler than the explanation that Dr. Weismann would substitute for it. The formation of the [canyon] of the Colorado by a sudden rift in the earthís crust, similar to those which opened during the Calabrian earthquakes, is simpler than its formation by the fretting of the stream during long ages under varying meteorological conditions. In these cases and in many others the simplest explanation is not the one accepted by science. Moreover, the simplicity of the explanation of the phenomena of animal activity as the result of intellectual processes, can only be adopted on the assumption of a correlative complexity in the mental nature of the animal as agent. And to assume this complexity of mental nature on grounds other than those of sound induction, is to depart from the methods of scientific procedure. [paragraph break] But what, it may be asked, is the logical basis upon which this principle is founded? (Morgan, 1894, pp. 54-55)
Obviously, the "logical basis upon which this principle is founded," was not on the basis of a simplicity criterion. Morgan showed in the preceding quotation that simplicity could be an untrustworthy and misleading criterion for choosing among alternative explanations. It is no wonder that Wozniak (1993), who criticized some of those who had misrepresented Morganís canon, wrote, "Even if one set out deliberately to distort the meaning of Morganís canon, it would be virtually impossible to do so with greater success. Morganís canon is not a principle of parsimony...." (p. ix-x)
Some of the history of the law or principle of parsimony will be addressed below, but it may be of some interest at this point to note that the Oxford English Dictionary, which includes, to the extent possible, quotations manifesting the first and other early published uses of each defined term or phrase, cited Sir William Hamilton as the first documented use of "law of parsimony" (see more on Hamilton below), and Morganís Animal Life and Intelligence (1890) was cited as the third documented use of the phrase, "law of parsimony." The quotation from Morgan clearly bears on his views on parsimony, "We do not know enough about the causes of variation to be rigidly bound by the law of parcimony." (p. 174; Hamilton and Morgan spelled it "parcimony") "Parcimony" appears in the index of Animal life and Intelligence (1890), but neither it, "Ockhamís razor," nor "parsimony" appeared in the indexes of Morganís An Introduction to Comparative Psychology (1894) or Animal Behaviour (1900). It may be pertinent to note that Animal Behaviour began as a revision of Animal life and Intelligence, but when "...it appeared that the amended treatment would not fall conveniently under the previous scheme of arrangement. I, therefore, decided to write a new book...." (Morgan, 1900, Preface)
Morganís canon in relation to anthropomorphism. Turning now to, perhaps, the second most frequent misinterpretation, namely, that the canon was anti-anthropomorphic. Consider how Morgan addressed the question of how one was one supposed to assess the lower and higher processes. According to Morgan (1894),
As we have already seen, we are forced, as men, to gauge the psychical level of the animal in terms of the only mind of which we have first hand knowledge, namely the human mind. But how are we to apply the gauge? (p. 55)
Morgan explained how to apply the gauge on pages 56-59, and it clearly involves anthropomorphic reasoning by analogy; see also, Costall (1993, 1998, and selected quotations in Table 3). Burghardt (1985, pp. 911-913) provided a useful account of how Morgan proposed that humans must use introspection to study animal minds. Wozniak (1993), who was quoted above on the point of misrepresentation of the canon by association with parsimony, also addressed those who misrepresented it as being anti-anthropomorphic. He said, "Even worse, it [Morganís canon] is consciously anthropomorphic and based squarely on the adequacy of the psychologistís personal introspection." (Wozniak, 1993, pp, ix-x)
Ockhamís Razor, the Law of Parsimony, and Simplicity
Because the most frequent misrepresentations of Morganís canon are linked to the simplicity criterion, typically, by referring to Morganís canon as being animal psychologyís closely related if not equivalent version of Ockham's razor and/or the law or parsimony, it will be useful to review briefly the origin and intent of Ockham's razor and the law of parsimony. Additionally, because Morganís canon has continued to be represented as manifesting a simplicity criterion for choosing among alternative explanations regarding animal behavior (e.g., Dewsbury, 2000; see quotation in Table 2, and see Table 4) whether or not Ockham's razor or the law of parsimony were invoked, it may be useful to address briefly some contemporary views regarding simplicity criteria per se in the sciences.
Ockhamís razor. Scholars who specialize in William Ockham (c.1285-1349) tend to spell his surname "Ockham" (e.g., Adams, 1987; Buytaert, 1958; Maurer, 1999; Moody, 1967). However, the Oxford English Dictionary uses "Occam" as the preferred spelling, and most contemporary writers seem to prefer Occam. Ockham will be used here except when someone is being quoted who used Occam.
William of Ockham has been described as "...the most influential philosopher of the fourteenth century...." (Moody, 1967, p. 306). Ockhamís razor, a relatively minor part of his body of work, was a methodological stricture for choosing among alternative logical formulations. While the most quoted version of Ockhamís razor appears to be, Entia non sunt multiplicanda praeter necessitatem, (entities are not to be multiplied without necessity; Moodyís translation) apparently, this version has not been found to be among Ockhamís writings (Burns, 1915; Moody, 1967; Thornburn, 1915). However, several apparently equivalent or closely comparable statements attributed to Ockham are known (see Adams, 1987 for several documented versions of Ockhamís razor). To cite two examples as translated by Moody (1967), "Plurality is not to be assumed without necessity" and "What can be done with fewer [assumptions] is done in vain with more." (p.307). For more on Ockhamís razor and its predecessors, see Adams (1987), Burns (1915), Buytaert (1958), Thornburn (1915), and Maurer (1999).
Ockham scholars do not always agree regarding the applicability of Ockhamís razor. For example, in the context of commenting on the apparent lack of occurrence in Ockhamís writings of entia non sunt multiplicanda necessitatem, Burns (1915) wrote:
The force of Ockhamís objection against Scotus was that logic and metaphysics were distinct. Both the thing and the universal are "entia," one "in re" and the other "in mente". Only a Scotist could think that the law of parcimony had anything to do with "entia." This is perhaps a mere matter of words; but words to a man like Ockham were not unimportant, and he was very careful with his original razor to make it cut only hypotheses (ponere, etc.). As a hit at a Scotist, he might have said "You must not make so many realities"; but in his philosophical argument he never seems to have forgotten that "entia" are quite untouched by logic. (p. 592)
However, Adams (1987) cited four versions of Ockhamís razor (identified as "D" through "G," including two that were similar to those translated by Moody above (1967), about which she concluded:
They are in the first instance, methodological principles, and it is not obvious how they are related to truth or even to probability. As such, they could serve as the basis of pragmatic arguments about what it would be futile or superfluous to do, or what one ought to do, but not of demonstrations in speculative metaphysics about what entities really exist. In this, (D) - (G) contrast with
(H) No plurality exists which cannot be proved from reason, experience or infallible authority,
(J) God never does with more what He could do with fewer.
Which could serve as premises in valid deductive arguments concerning what entities really exist. (p. 158).
Finally, for present purposes, Maurer (1999) wrote:
We can be certain that he [Ockham] regarded the razor as a principle, not in demonstrative but in dialectical reasoning, and as such it plays a significant role in shaping his views on reality and the mind. (p. 129)
Lest one read too much into the latter clause, a careful reading of Maurer (pp. 112-129) is recommended.
Maurer (1999) also noted that Walter Chatton, a contemporary critic of Ockham, proposed an anti-razor, "My rule...if three things are not enough to verify an affirmative proposition about things, a fourth must be added, and so on." (Maurer, 1999, p. 127, and see Boring, 1929 and 1950 in Table 3 of the present work). Maurer also recounted some of the debate between Chatton and Ockham concerning the razor. In any case, whether or not Ockhamís razor might be applied appropriately to the kinds of natural phenomena that Morganís canon addressed, Hamilton (see next section) made it clear that his "law of parsimony" was intended to address natural phenomena.
Law of parsimony. "Occam's" razor was equated to the "Law of Parcimony" in the Oxford English Dictionary (Simpson & Weiner, 1989, Volume XIII, p. 245). However, the "law of parsimony" appears to be most associated with Sir William Hamilton (1788-1856). Hamilton has been recognized for extending the applicability of Ockhamís razor to explanations associated with natural phenomena. According to Pearson (1892), Sir William Hamilton "...in a valuable historical note..." extended Occamís razor as follows. First, Hamilton extended it with some "further scholastic axioms" (see Pearson, p. 482). Pearson assessed the result of that extension as: "So far these axioms are valuable as canons of thought, they express no dogma but a fundamental principle of the economy of thought." (p. 482). Pearson continued:
When, however, Sir William Hamilton adds to them, Natura horret superfluum, and says that they only embody Aristotle and Newtonís dicta that God and Nature never operate superfluously and always through one rather than a plurality of causes, then it seems to me we are passing from the safe field of scientific thought to a region thickly strewn with the pitfalls of metaphysical dogma. (p. 482)
After addressing some of the pitfalls, Pearson continued:
Sir William Hamilton expresses Occamís canon in the more complete and adequate form: - Neither more, nor more onerous causes are to be assumed than are necessary to account for the phenomena. (p. 482)
Pearson cited Hamiltonís "Discussions on Philosophy, 2nd edition, pp. 628-31, London, 1853," which was not readily available to the present writer. However, Hamilton can be quoted directly on the "law of parcimony" from sources that were available as follows:
Without descending to details...there exists a primary presumption of philosophy. This is the law of parsimony; which prohibits, without a proven necessity, the multiplication of entities, powers, principles or causes; above all, the postulation of an unknown force where a known impotence can account for the phenomenon. We are therefore entitled to apply "Occamís razor" to this theory of causality, unless it be proved possible to explain the causal judgment at a cheaper rate...." (Hamilton, 1855, p. 580; a slightly, differently worded version of the quotation may be seen in Hamilton, 1869, Volume 1, p. 586.)
Simplicity as a criterion in contemporary science. The ability and desirability to choose among alternative explanations in science, including behavioral science, based on choosing the simpler or simplest explanation appears to be taken for granted by most psychologists who invoke the criterion. However, philosophers of science who have considered the simplicity criterion have raised serious doubts regarding its feasibility in most circumstances. For example, see Bungeís (1963), Myth of simplicity: Problems of scientific philosophy, or for more recent accounts, see several authors included in Boyd, Gaspar, and Trout (1991; refer to "simplicity" in the index).
While this is not the place, nor am I prepared to try, to summarize all the important considerations associated with the general impracticality or fallibility of a simplicity criterion, one important consideration that is likely to be applicable to any attempt to apply a simplicity criterion to the kinds of explanations of phenomena addressed by Morganís canon is the fallibility of the words in which explanations are expressed or the fallibility of understanding all that may be applicable to each potential explanation. For example, a seemingly simpler explanation that may seem to embody one or only a few assumptions may embody many hidden or unrealized assumptions. This may have been part of what Morgan was trying to demonstrate with his examples, quoted earlier, to show his misgivings about simplicity as a criterion. Whether it was his intention or not, please reconsider Morganís example regarding the formation of the Colorado canyon, where formation by earthquake was offered as a "simpler" explanation compared to the "...fretting of the stream during long periods of varying meteorological conditions" (Morgan, 1894, p. 54). According to Morgan, earthquake as a cause for the canyon was the simpler but was not the scientifically accepted explanation. Nevertheless, that seemingly simpler earthquake explanation, in light of modern knowledge, overlooks the complex chain of geological events that likely also occurred "during long periods" that result in an earthquake.
Please see, also, a brief and equally skeptical, discussion of the simplicity criterion as it might be applied to research in animal cognition and, thus, Morganís canon, in Thomas (1998). Finally, on the point of simplicity as a criterion for choosing among scientific explanations, in his recently published Dictionary of Philosophy, Bunge (1999) defined "simplism" as:
The view that the simplest hypotheses, theories, or methods are always to be preferred. Given the complexity of the real world, the only justification for simplism is laziness. Indeed the history of knowledge is, on the whole, one of increasing complexity. (p. 269)
A History of Misinterpretation of Morganís Canon and Efforts to Correct the Misrepresentations
With the preceding sections of this paper as background, central to this section and to this article are Table 2 and Table 3. Table 2 includes quotations illustrative of the history of misinterpretation and misrepresentation of Morganís canon. Table 3 includes quotations illustrative of the history of attempts to correct some of the misinterpretations and misrepresentations. Before examining Tables 2 and 3, it is suggested that the following paragraph be read as a preface to Table 2.
Regarding Table 2, considerable effort was made to avoid misrepresenting an author's intentions, for example, by obscuring the context. Additionally, because some of the quotations in Table 2 may appear to be sufficiently well qualified, provided one knew any of the history associated with the misrepresentation of Morgan's canon, the following criterion was used to decide whether an author should be included in Table 2. It was deemed appropriate to include an author in Table 2 when that author associated Morganís canon with Ockhamís razor, the law of parsimony, or a simplicity criterion or represented Morgan's canon as being anti-anthropomorphic or anti-anecdotal, if that author did so without negating the association with regards to Morgan's intentions or did so without acknowledging any of the history of misrepresentation in conjunction with such associations. In other words, in the light of the many efforts to correct the misrepresentation of Morgan's canon, to be less than clear when referring to Morgan's canon in conjunction with any of the known misrepresentations was deemed to continue the misrepresentation of Morgan's canon. The reader is urged to read the original sources and judge for her/himself whether the selected quotations accurately represent the authorsí intentions and whether the quotations contribute to the misrepresentation of Morgan's canon. Finally, rather than try to include all pertinent authors and references in Table 2 and Table 3 (a task likely to be near-impossible for Table 2), the sources used were selected, either by convenience (viz., articles and books I had already collected over the years) or in pursuit of answers to specific questions. It is highly likely that many more examples of misrepresentation could be found.
In some cases, an author who misrepresented Morganís canon in the ways quoted in Table 2 may also have addressed it correctly (or incorrectly) in other ways not quoted there (e.g., Boakes, 1984, Dewsbury, 1984). However, except for Adams (1928), who was included for his early involvement, and Boring (1929, 1950), who was included for the impact he likely had on many psychologistsí understanding of Morganís canon, it was deemed likely to be too tedious to try to sort the correct use from the misrepresentation in most such cases. As for Boakes and Dewsbury, the reader is urged to read them for her/ himself, and see the next paragraph here. Except, perhaps, for the lapse quoted in Table 2, Boakes's historical account of Romanes, Morgan, and their relationship is a valuable contribution to the Morgan-Romanes literature.
In fairness to Dewsbury, the quotations attributed to him in Table 2 for the years 1973 and 1978 occurred before he was on record (Dewsbury, 1979) as having read Morganís An Introduction to Comparative Psychology (1894). Dewsbury (1984), presumably, used that experience and also drew on Newbury (1954; see Table 3 here) and others (e.g., Gray, 1963; Nagge, 1932; Table 3) who had pointed to the misrepresentation of Morganís canon to write the following:
In the prevailing view, Morganís canon is essentially equivalent to the "law of parsimony" or "Occamís razor." My first teacher in the history of psychology, P. L. Harriman, termed Morganís canon "a restatement of the principle, expounded by William of Occam" (1947, 225-226; see also 255). I followed Harriman and others (1973a, 1978a). This appears generally inaccurate." (Dewsbury, 1984, p. 187)
Although he now disagreed with Harriman regarding Morgan's canon and Ockham's razor, nevertheless Dewsbury (1984) wrote, "The law of parsimony and Morgan's canon are two closely related principles." (p. 188) Please see also the inclusion of Harriman (1947) in Table 2, where he equated Morgan's canon with Ockham's razor and with the law of parsimony, and see Wozniak (1993) in Table 3 who strongly criticized Harriman for that misrepresentation.
In addition to those who misrepresented Morgan's canon and those who tried to correct the misrepresentation, there is, of course, a third group of authors, not included systematically here, who have used Morganís canon mostly or entirely appropriately or who have discussed its implications or weaknesses appropriately but did not address the misrepresentation issue. No effort was made to include these here except to cite a couple of examples now. For example, Roitblatís (1987) discussion of Morganís canon on pages 31-32 seems appropriate, although he did appear to link it with Occamís razor on page 296. Robertsí (1998) discussion of Morganís canon on pages 7-9 seems exceptionally well done, although he, too, linked the canon to the simplicity criterion (see last whole paragraph on page 8). In any case, additions to the references cited in this paper and comments about it are encouraged.
How Morganís Canon is Represented in Recent History of Psychology Textbooks
Table 4 summarizes the representations and misrepresentations of Morganís canon in 17 recent history of psychology textbooks spanning the years 1991-2001 . These textbooks are in my personal library, and that was the only basis by which they were chosen for this survey. They were examined in terms of how they represented Morgan's canon in conjunction with one or more of the following: (1) the law of parsimony, (2) Ockham's razor, (3) a simplicity criterion, (4) anthropomorphism, and (5) use of anecdotes. While in most cases, the authorsí interpretations in terms of these categories were sufficiently clear, in a few cases, they were not. Again, readers are urged to read the sources cited and decide for themselves whether Table 4 accurately represents the textbooks authorsí views. The "richness" of misrepresentation is not always reflected by the table. For example, Goodwin (1999) headed the section that included coverage of Morganís canon with the large-type, bold-print, "CONWY LLOYD MORGAN (1852-1936): THE PRINCIPLE OF PARSIMONY."
Additionally, the table does not reflect that some writers had a generally appropriate discussion of Morganís canon that was, perhaps, only slightly marred by misrepresentation. For example, Benjafieldís (1996) discussion of Morganís canon drew on Costall (1993; see Table 3) and reflected Costallís views well on the point that Morganís canon was not anti-anthropomorphic; however, Benjafield also wrote that the canon "urged" explanation in terms of "simpler cognitive processes."
In any case, 14 of the 16 textbooks (counting Leahey 2000 and 2001 as one for present purposes) that were examined addressed Morganís canon, which suggests that it is a sufficiently important topic to include in most history of psychology textbooks. However, to a greater or lesser extent, 10 of the 14 textbooks misrepresented Morganís canon (or 11, depending on how one views Watson & Evans, 1991; see footnote 3 in Table 4). It is somewhat discouraging that the textbooks currently used to teach history of psychology would show such a high proportion of misrepresentation of a topic covered by so many of them. However, Bolles (1993; now deceased), Smith (1997) and Thorne and Henley (2001) should be commended for their coverage of Morganís canon. An encouraging note is that Costallís article (1993; and see Table 3) is having a good influence as exemplified by Thorne and Henleyís coverage and most of Benjafieldís (1996).
Elsewhere, Thomas (1997a; updated here to include all 17 textbooks cited in Table 4) has reported a similar survey of the representation in history of psychology textbooks of another topic, the "discovery of the speech center." That survey revealed that 13 of 16 (again, counting Leahey, 2000/2001 as one) addressed the topic and that 9 of the 13 misrepresented it as follows. The nine attributed the discovery to Pierre Paul Broca (1824-1880) alone, ignoring others such as Simon Alexandre Ernest Auburtin (1825-1893) and Jean Baptiste Bouillaud (1796-1875) who deserve as much or more credit than did Broca. Representing it appropriately among history of psychology textbooks were Benjamin (1997), Fancher (1996), Hothersall (1995) and Thorne & Henley (2001). The degree of misrepresentation in history of psychology textbooks on this topic may be contrasted with its coverage in contemporary histories of neuroscience where 4 of 4 represented it correctly (Finger, 1994; Marshall & Magoun, 1998; Plum & Volpe, 1987; Young, 1970/1990).
Considering the examples of misrepresentation of Morganís canon and of the discovery of the speech center (see also the misrepresentation of Pavlovís "mugging;" Thomas, 1997b) in history of psychology textbooks, one wonders how they might stand up to comparable assessments on other topics. I hasten to say it that takes considerable desire and stamina to attempt to write a history of psychology. Those who do it reasonably well have my admiration and respect.
Morganís revisionist views. As the years passed, Morgan changed some of his earlier positions, and these changes may have contributed to the misrepresentation of Morganís canon. For example, in Animal Behaviour, Morgan (1900) appeared to link simple and complex to lower and higher processes, respectively. In the context of considering conscious control and the role of memory in the behavior of a chick, Morgan wrote:
It may still be said, however, that in selecting an example from so highly organized an animal as a bird, we are taking for granted that a complex case of controlled behavior may fairly be accepted as a type of more simple cases. Unfortunately, the only being with whose power of conscious control we have any first hand acquaintance is possessed of an nervous system even more organized than that of a chick. Our psychological interpretations are inevitably anthropomorphic. All we can hope to do is to reduce our anthropomorphic conclusions to their simplest expression." (p. 48).
Thus, while this might seem to favor simple over complex explanations or to link higher and lower processes to complex and simple, respectively, precisely what this meant is unclear in terms of advocating a simplicity criterion, and it has no directly meaningful connection to Ockhamís razor or the law of parsimony. Furthermore, who can say what "reduce our anthropomorphic conclusions to their simplest expression" means?
Regarding his view of the value of anecdotes, Morgan clearly revised his opinion as the following quotations will demonstrate. First, consider what he wrote just prior to the canon and concurrently with the canon. In 1890, Morgan wrote:
I do not propose to bring forward a number of new observations on the highly intelligent actions which animals are capable of performing. Mr. Romanes has given us a most valuable collection of anecdotes on the subject in his volume on "Animal Intelligence." (p. 362)
Consistent with this, a portion of Morganís eulogy for Romanes (as quoted in Sandersonís, 1895, tribute to Romanes in the Proceedings of the Royal Society of London; see also E. Romanes, 1902, p. 202):
...by his patient collection of data, by his careful discussion of these data in the light of principles clearly formulated; by his wide and forcible advocacy of his views, and above all by his own observations and experiments, Mr. Romanes left a mark in this field of investigation and interpretation which is not likely to be effaced. (Quoted in Sanderson, 1895, p. xiii)
This was being said about the time An Introduction to Comparative Psychology (1894) and Morganís canon were being published. Morgan acknowledged Romanesí death in a footnote that was added to the bookís Preface after the Preface had been written. The phrase above, "discussion of these data in the light of principles clearly formulated," strongly suggests that Morganís reference was meant to include if not emphasize the anecdotal data (see more below in conjunction with "principles clearly formulated").
However, by 1897, Morgan had begun to question Romanesí use of anecdotal data. Morganís (1897) biographical sketch of Romanes written for the Dictionary of National Biography included:
In 1881 [sic] he published in the ĎInternational Scientific Seriesí under the same title [Animal Intelligence] that he had given to his Dublin lecture, a collection of data, perhaps too largely anecdotal, respecting the mental faculties of animals in relation to those of man. (Volume XLIX, p. 178)
In his autobiographical chapter, Morgan (1932) wrote:
With regard to Romanes collection of anecdotes, psychologically interesting in its way, I felt, as no doubt he did, that not on such anecdotal foundations could a science of comparative psychology be built. Most of the stories were merely casual records, supplemented by amateurish opinions of passing observers who psychological training was well-nigh negligible. (p. 247)
Particularly related to "principles clearly formulated" by Romanes as well as the second sentence in the preceding quotation and in defense of Romanes, it may be noted that both the principles and most of the anecdotes that Romanes used as data appeared in his book, Animal Intelligence (1882). Romanes made a considerable effort in the bookís preface to show his concern about the use of anecdotes, and he stated several principles regarding collecting them, presenting them, and evaluating them. On the point of presenting the anecdotes, he apparently felt an obligation to quote them verbatim as much as possible and to abbreviate them only when necessary and as little as possible (see p. xi). The apparent sense of duty to the original observer probably tarnished Romanes reputation a great deal.
For example, Romanes was belittled by no less than Wilhelm Wundt (1832-1920; see Wundt, 1896/1901) and Margaret Washburn (1871-1939; see Washburn,1908) for an anecdote associated with the "funereal habits" of ants. Yet, there was no evidence that Romanes had accepted as valid the original observersí interpretations. Romanes quoted the anecdotes, presumably, to establish the fact that ants may bury other ants. He provided a reasonable and conservative interpretation of of his own to explain why ants may bury dead ants. His interpretation was based on natural selection associated with the sanitation, thus health, of the colony (see Romanes, 1882, p. 89), but his interpretation was apparently overlooked by his critics .
What may have prompted Morganís changing views regarding the value of the anecdotes has not been determined (by me), but at least one of his contemporaries T. Wesley Mills (1847-1915; see Mills,1899) offered an uncharitable opinion. Mills, whose approach to comparative psychology may be compared to Romanesís and Morganís at the time and contrasted with E. L. Thorndikeís (1874-1949; e.g., Thorndike, 1898) experimental method, became engaged in published debate with Thorndike. In this context, on one occasion Mills wrote:
But Professor Morgan is more and more in sympathy with the destructive school [an apparent reference to Thorndike], so that he now seems willing to surrender anything to all and sundry who may ask him to stand and deliver. (Mills, 1899, p. 271)
Please see Costall (1993, 1998) or Gray (1963b) for further information pertinent to the issue of why Morganís canon should not have been interpreted as being anti-anecdotal nor anti-Romanes.
In any case, Morganís views appeared to have undergone some revision, as would be reasonable for any of the participants in this early stage of the development of comparative psychology. Nevertheless, when evaluating the representation of Morganís canon and when citing 1894 as the source, to represent the canon accurately, one must represent Morganís views when he wrote it.
Morganís canon and the changing Zeitgeist. Although this topic will not be developed here, the persistent misrepresentation of Morganís canon as a canon of parsimony and as being anti-anthropomorphic fit the Ďspirit of the timesí when psychology was struggling to be recognized as an experimental science, and the misrepresented view of Morganís canon fit well with the emergence of behaviorism. With the re-assertion of "cognitive psychology" in the 1960s and its concomitant, "animal cognition," there has been a renewed interest in and efforts to justify what has been termed "the new anthropomorhism" (Kennedy, 1992; see also Mitchell, Thompson, & Miles, 1997). Morganís original intentions regarding the canon fit very well with the contemporary Zeitgeist.
Adams, D. K. (1928). The inference of mind. Psychological Review, 35, 235-252.
Adams, M. M. (1987). William Ockham: Volume 1. Notre Dame, IN: University of Notre Dame Press.
Baenninger, R. (1994). A retreat before the canon of parsimony. Contemporary Psychology, 39, 505-507.
Barrows, E. M. (1995). Animal behavior desk reference. Boca Raton: CRC Press.
Battig, W. F. (1962). Parsimony in psychology. Psychological Reports, 11, 555-572.
Bekoff, M, & Allen, C. (1997). Cognitive ethology: Slayers, skeptics, and proponents. In R. W. Mitchell, N. S. Thompson, & H. L. Miles (Eds.). Anthropomorphism, anecdotes, and animals (pp. 313-334). Albany, NY: State University of New York Press.
Benjafield, J. G. (1996). A history of psychology. Boston: Allyn and Bacon.
Benjamin, Jr., L. T. (1997). A history of psychology: Original sources and contemporary research (2nd ed.). New York: McGraw-Hill.
Boakes, R. (1984). From Darwin to behaviorism: psychology and the minds of other animals. London: Cambridge University Press.
Bolles, R. C. (1993). The story of psychology: A thematic history. Pacific Grove, CA:
Boring, E. G. (1929). A history of experimental psychology. New York: The Century Co.
Boring, E. G. (1950). A history of experimental psychology: (2nd ed.). New York: The Century Co.
Boyd, R., Gasper, P., & Trout, J. D. (Eds.) (1991). The philosophy of science. Cambridge, MA: The MIT Press.
Brennan, J. F. (1998). History and systems of psychology (5th ed.). Englewood Cliffs, NJ: Prentice Hall.
Burghardt, G. M. (1985). Animal awareness: Current perceptions and historical perspective. American Psychologist, 40, 905-919.
Burns, C. D. (1915). Occamís razor. Mind, 24, 592.
Buytaert, E. M. (Ed.) (1958). Philotheus Boehner, O.F.M.., Ph.D.: Collected articles on Ockham. St. Bonaventure, NY: Franciscan Institute Publications.
Caldwell, W. E. (1960). Theoretical foundations of comparative psychology. In R. H. Waters, D. A. Rethlingshafer, & W. E. Caldwell (Eds.). Principles of comparative psychology (pp. 378-404). New York: McGraw-Hill.
Costall, A. (1993). How Lloyd Morganís Canon backfired. Journal of the History of the Behavioral Sciences, 29, 113-122.
Costall, A. (1998). Lloyd Morgan, and the rise and fall of "animal psychology." Society and Animals, 6, 13-29.
Costall, A., Clark, J.F.M. & Wozniak, R. H. (1997). Conwy Lloyd Morgan (1852-1936): An introduction to his work and a bibliography of his writings. Teorie & Modelli, 2, 65-92.
Denny, M. R. (1980). Comparative psychology: An evolutionary analysis of behavior. New York: John Wiley & Sons.
Dewsbury, D. A. (1973). Introduction. In D. A. Dewsbury, & D. A. Rethlingshafer (Eds.). Comparative psychology: A modern survey (pp. 1-18). New York: McGraw-Hill.
Dewsbury, D. A. (1978). Comparative animal behavior. New York: McGraw-Hill.
Dewsbury, D. A. (1979). C. Lloyd Morgan: Something old thatís often new. Contemporary Psychology, 24, 677-680.
Dewsbury, D. A. (1984). Comparative psychology in the twentieth century. Stroudsburg PA: Hutchinsin Ross Publishing Co.
Dewsbury, D. A. (1993). On publishing controversy: Norman R. F. Maier and the genesis of seizures. American Psychologist, 48, 869-877.
Dewsbury, D. A. (2000). Issues in comparative psychology at the dawn of the 20th century. American Psychologist, 55, 750-753.
Dixon, E. T. (1892). "The Limits of Animal Intelligence." Nature, 46, 392-393.
Epstein, R. (1984). The principle of parsimony: Some applications in psychology. The Journal of Mind and Behavior, 5, 119-130.
Fancher, R. E. (1996). Pioneers of psychology (3rd ed.). New York: W.W. Norton.
Finger, S. (1994). Origins of neuroscience: A history of explorations of brain function. New York: Oxford University Press.
Flugel, J. C. (1933). A hundred years on psychology. London: Gerald Duckworth & Co.
Goodwin, C. J. (1999). A history of modern psychology. New York; John Wiley & Sons.
Gray, P. H. (1963a). Morganís Canon: A myth in the history of comparative psychology. Proceedings of the Montana Academy of Sciences, 23, 219-224.
Gray, P. H. (1963b). The Morgan-Romanes controversy: A contradiction in the history of comparative psychology. Proceedings of the Montana Academy of Sciences, 23, 225-230.
Grier, J. W., & Burk, T. (1992). Biology of animal behavior (2nd ed.). St. Louis, MO: Mosby Year Book.
Griffin, D. R. (1981). The question of animal awareness. New York: The Rockefeller University Press.
Guthrie, R. V. (1998). Even the rat was white. Boston: Allyn & Bacon.
Hamilton, W. (1855). Discussions on philosophy and literature. New York: Harper & Brothers, Publishers.
Hamilton, W. (1869). Lectures on metaphysics and logic. Volume I: Metaphysics. (Edited by H. L. Mansel & J. Veitch). Boston: Gould and Lincoln.
Harriman, P. L. (Ed.) (1947). New dictionary of psychology. New York: Philosophical Library.
Hergenhahn, B. R. (2001). An introduction to the history of psychology (4th ed.). Pacific Grove, CA: Brooks/Cole.
Holmes, S. J. (1911). The evolution of animal intelligence. New York: Henry Holt and Company.
Hothersall, D. (1995). History of psychology (3rd ed.). New York: McGraw-Hill.
Hunt, M. (1993). The story of psychology. New York: Doubleday.
Kennedy, J. S. (1992). The new anthropomorphism. Cambridge: Cambridge University Press.
Knoll, E. (1997). Dogs, Darwinism, and English sensibilities. In R. W. Mitchell, N. S. Thompson, & H. L. Miles (Eds.). Anthropomorphism, anecdotes, and animals (pp. 12-21). Albany, NY: State University of New York Press.
Leahey, T. H. (2000). A history of psychology: Main currents in psychological thought (5th ed.). Upper Saddle River, NJ: Prentice Hall.
Leahey, T. H. (2001). A history of modern psychology (3rd ed.). Upper Saddle River, NJ: Prentice Hall.
Macphail, E. M. (1998). The evolution of consciousness. New York: Oxford University Press.
Marshall, L. H., & Magoun, H. W. (1998). Discoveries in the human brain. Totowa,NJ: Humana Press.
Maurer, A. (1999). The philosophy of William of Ockham in the light of its principles. Toronto: pontifical Institute of Mediaeval Studies.
Miller, G. A. (1962). Psychology: The science of mental life. New York: Harper & Row.
Mills, W. (1899). The nature of animal intelligence and the methods of investigation it. Psychological Review, 6, 262-276.
Mitchell, R. W., Thompson, N. S., & Miles, H. L. (Eds.) (1997). Anthropomorphism, anecdotes, and animals. Albany, NY: State University of New York Press.
Moody, E. A. (1967). William of Ockham. In P. Edwards (Ed.). The encyclopedia of philosophy: Volume 8. (pp. 306-317). New York: Macmillan Publishing Co., Inc.
Morgan, C. L. (1890). Animal life and intelligence. London: Edward Arnold.
Morgan, C. L. (1894). An introduction to comparative psychology. London: Walter Scott, Ltd.
Morgan, C. L. (1897). Romanes, George John (1848-1894). Dictionary of National Biography, Volume LVII (pp. 177-180). New York: The Macmillan Co.
Morgan, C. L. (1900). Animal behaviour. London: Edward Arnold.
Morgan, C. L. (1903). An introduction to comparative psychology (New ed., revised). London: The Walter Scott Publishing Co. Ltd.
Munn, N. L. (1950). Handbook of psychological research on the rat. Boston: Houghton Mifflin Co.
Nagge, J. W. (1932). Regarding the law of parsimony. Journal of Genetic Psychology, 41, 492-494.
Newbury, E. (1954). Current interpretations and significance of Lloyd Morganís Canon.
Psychological Bulletin, 51, 70-74.
Pearson, K. (1892). The grammar of science. London: Walter Scott.
Pillsbury, W. B. (1929). The history of psychology. New York: W. W. Norton & Co.
Plum, F., & Volpe, B.T. (1987). Neuroscience and brain function: From myth to public responsibility. In V.B. Mountcastle, F. Plum & S. R. Geiger (Eds.), Handbook of physiology: Section 1: The nervous system. Bethesda, MD: American Physiological Society.
Richards, R. J. (1987). Darwin and the emergence of evolutionary theories of the mind and behavior. Chicago: University of Chicago Press.
Roberts, W. A. (1998). Principles of animal cognition. New York: McGraw-Hill.
Roberts, W. H. (1929). A note on anthropomorphism. Psychological Review, 36, 95-96.
Roitblat, H. L. (1987). Introduction to comparative cognition. New York: W. H. Freeman, Co.
Romanes, E. (1902). The life and letters of George John Romanes (5th ed.). London: Longmans, Green, and Co.
Romanes, G. J. (1882). Animal intelligence. London : K. Paul, Trench.
Romanes, G. J. (1883). Mental evolution in animals. London : K. Paul, Trench.
Romanes, G. J. (1888). Mental evolution in man. London : K. Paul, Trench.
Sanderson, J. B. (1895). George John Romanes. Proceedings of the Royal Society of London, 57, vii-ivx.
Schultz, D.P., & Schultz, S.E. (2000). A history of modern psychology (7th ed.). Fort Worth, TX: Harcourt Brace College Publishers.
Singer, B. (1981). History of the study of animal behaviour. In D. McFarland (Ed.). The Oxford companion to animal behaviour. (pp. 255-272). Oxford: Oxford University Press.
Smith, R. (1997). The human sciences. New York: W. W. Norton & Co.
Stanley, H. M. (1896). Remarks on Professor Lloyd Morganís method in animal psychology. Psychological Review, 3, 536-541.
Thomas, R.K. (1997a, June). Two contrasting views regarding the discovery of the speech center and recognition of its importance for the theory of localization of brain function. Poster presented at the meeting of Cheiron: The International Society for the History of the Behavioral and Social Sciences, Richmond, VA.
Thomas, R. K. (1997b). Correcting some Pavloviana regarding "Pavlovís Bell" and Pavlovís "mugging." American Journal of Psychology, 110, 115-125.
Thomas, R. K. (1998). Lloyd Morganís Canon. In G. Greenberg & M. M. Haraway (Eds.). Comparative psychology: A handbook. (pp. 156-163). New York: Garland Press.
Thornburn, W. M. (1915). Occamís razor. Mind, 24, 287-288.
Thorne, B. M., & Henley, T. B. (2001). Connections in the history and systems of psychology. Boston: Houghton Mifflin. Co.
Thorndike, E. L. (1898). Animal intelligence: An experimental study of the associative processes in animals. The Psychological Review: Monograph Supplements, II, No. 4 (Whole No. 8). 1-109.
Viney, W., & King, D. B. (1998). A history of psychology. Boston: Allyn & Bacon.
Walker, S. (1983). Animal thought. London: Routledge and Keegan Paul.
Washburn, M. F. (1908). The animal mind: A text-book of comparative psychology. New York: The Macmillan Co.
Warden, C. J. (1927). The historical development of comparative psychology. (IV. From the sixteenth century to Darwin. V. The modern experimental movement.). Psychological Review, 34, 135-168.
Waters, R. H. (1939). Morganís Canon and anthropomoprhism. Psychological Review, 46, 534-540.
Waters, R. H., Rethlingshafer, D. A., & Caldwell, W. E. (1960). Principles of comparative psychology. New York: McGraw-Hill.
Watson, Sr., R. I., & Evans, R. B. (1991). The great psychologists: A history of psychological thought (5th ed.). New York: HarperCollins Publishers.
Wozniak, R. H. (1993). Conwy Lloyd Morgan, mental Evolution, and the Introduction to comparative psychology: An introduction. (Introduction to re-publication). Morgan, C. L. Introduction to comparative psychology. London: Routledge/Thoemmes Press. (Original work published 1894)
Wundt, W. (1896/1901). Lectures on human and animal psychology, 2nd ed..
Young, R. M. (1970/1990). Mind, brain, and adaptation in the nineteenth century. New York: Oxford University Press.
Selected Examples in Chronological Order to Illustrate the Histories of (1) Misrepresentations of Morganís Canon; see also, Table 4, and (2) Efforts to Correct the Misrepresentations of Morganís Canon.
Misrepresentations Efforts to Correct
Warden, 1927 Adams, 1928
Adams, 1928 Boring, 1929
Flugel, 1933 Nagge, 1932
Munn, 1950 Newbury,1954
Caldwell, 1960 Miller, 1962
Denny, 1980 Singer, 1981
Griffin, 1981 Burghardt, 1985
Grier & Burke, 1992 Costall 1993
Baenninger, 1994 Wozniak, 1993
Barrow, 1995 Costall et al., 1997
Bekoff & Allen, 1997 Costall, 1998
Knoll, 1997 Thomas, 1998
See, also, Examples in Table 4
See, also, Examples in Table 4
Selected Quotations from the Sources Indicated in Table 1 to Illustrate the History of Misinterpretations and Misrepresentations of Morganís Canon. Sources and Quotations Are in Bold Print. Explanatory or supplementary comments appear in non-bold print.
Because a few of the quotations included here may seem questionable to some readers, it was deemed appropriate to include a quotation in Table 2 when its author associated Morganís canon with Ockhamís razor, the law of parsimony, or a simplicity criterion or when that author represented Morgan's canon as being anti-anthropomorphic or anti-anecdotal, if the author did so without negating the association with regards to Morgan's intentions or did so without acknowledging any of the history of misrepresentation in conjunction with such associations. In other words, in the light of the many efforts to correct the misrepresentation of Morgan's canon, to be less than clear when referring to Morgan's canon in conjunction with any of the known misrepresentations was deemed to continue the misinterpretation and misrepresentation of Morgan's canon. _______________________________________________________________________________________________
Stanley (1896, p. 541)
His [Morganís] caution is also admirable, but we do not think the law of parsimony is positive proof as he seems to urge.
Mills (1899, p. 271)
Nor can I agree with those who maintain that we must always adopt the simplest explanation of an animalís action.
Washburn (1908, p. 25)
[Immediately after quoting Morganís canon...] In other words, when in doubt take the simpler interpretation.
Holmes (1911, p. 159)
...it is well in general to be guided by the principle enunciated by Lloyd Morgan, which is a sort of special case of the law of parsimony.
Warden (1927, p. 155)
The canon of Morgan...was an attack against the prevailing anthropomorphism....The canon is merely the law of parsimony applied to animal psychology.
Adams (1928, p. 241)
[Immediately after quoting Morganís canon...] This is plainly intended as an adaptation to the problems of animal psychology of the general Law of Parsimony....
Boring (1929, pp. 464-465)
[Morgan]...who undertook to offset the anthropomorphic tendency in the interpretation of the animal mind by an appeal to the Ďlaw of parsimony.í This law applied to animal psychology is often known as ĎLloyd Morganís canon.í
Pillsbury (1929, p. 283)
He [Morgan] is deserving of credit for urging what he calls the law of parsimony in the interpretation of mental phenomena in animals....
Flugel (1933, pp. 123-124)
The reaction started with Lloyd Morgan, who, in the nineties, endeavored to combat the dangers of the anecdotal method by the "law of parsimony", according to which we must always explain animal behaviour in terms of the simplest mental processes that will account for the facts.
Waters (1939, p. 534)
Morganís canon was offered as just such a check [against the use of anecdotes and anthropomorphism]. Its immediate effect was to outlaw at once any description of animal behavior as due to mental processes.
Harriman (1947, pp. 225-226; 255)
These definitions appeared in Harrimanís, The New Dictionary of Psychology.
Morganís canon: C. Lloyd Morganís axiom to the effect that the simplest explanation of all known facts is the best hypothesis or theory. It is a restatement of the principle expounded by William of Occam (c. 1325) and known as Occamís razor.
parsimony, law of: Lloyd Morganís statement (1900) that animal behavior should be described in the simplest possible terms. It is an application of Occamís razor to animal psychology. Occam (1280-1349) had said that entities should not be multiplied beyond necessity; and Morgan accepted this view, indicating that anecdotes, attribution of human mental activities to animals, and projection of introspections have no place in animal psychology.
Munn (1950, pp. 1-2)
Lloyd Morgan...advocated a curb on anthropomorphic speculation....His well known principle of parsimony for students of animal behavior read as follows:...
Caldwell (1960, p. 401)
Morgan gave comparative psychology his interpretation of the law of parsimony, which curbed the tendency of observers of animals to anthropomorphize.
Dewsbury (1973, p. 9)
He proposed a law which has been variously termed Occamís razor, the law of parsimony, and Lloyd Morganís canon....Lloyd Morganís canon seems applicable today. If alternative explanations appear truly equal, the simpler is to be preferred until data require postulation of more complex processes.
Dewsbury (1978, p. 10)
Morgan is best known for opposing unbridled anthropomorphism. According to the often-cited "law of parsimony" or "Lloyd Morganís canon...." The admonition that we should strive to accept the simpler of two equal alternative explanations is certainly good advice for many situations.
Denny (1980, p. 4)
C. Lloyd Morgan, author of the famous Canon of Parsimony, dealt explicitly with animal behavior.
Griffin (1981, p. 118)
This [Morganís canon] has been widely interpreted as requiring that complex functions should not be postulated if a simpler explanation will suffice. That is the widely accepted principle of parsimony...
Here and elsewhere (p. 99) Griffin accepted the interpretation that Morganís canon is a canon of parsimony. However, Griffin also accepted (see p. 131) Millerís view (1962; see Table 3) that the canon was not anti-anthropomorphic.
Boakes (1984, p. 40)
The canon can be seen as simply the application of the general law of parsimony to explanation of behavior. Nevertheless, Morgan did not justify it on these terms but on the grounds of evolutionary theory.
Dewsbury (1984, pp. 188)
The law of parsimony and Morganís canon are two closely related principles.
Epstein (1984, pp. 122-123)
[Epsteinís article included a section titled "Ockham and Morgan" in which the following may be seen.] Morgan was a British psychologist and biologist who, in An Introduction to Comparative Psychology, published in 1894, challenged the tendency of some naturalists of his day to attribute human characteristics to animals....Morgan was no less a mentalist than Romanes, but he took a more conservative stand. Just as evolution had produced organisms that varied from the simple, to the complex, he argued, so must it have produced minds that varied from the simple to the complex. It would therefore be presumptuous of us to infer higher mental activities in animals where simpler ones would do. He expressed this in his famous Canon, sometimes called the Canon of Parsimony.
Grier & Burk (1990, p. 52)
Among his other contributions, he rejected anecdotalism and undisciplined anthropomorphism in the interpretation of behavior in other animals. He called for a principle of theoretical parsimony (i.e., the simplest explanation) which became known as Morganís canon:....
Baenninger (1994, p. 805)
Baenningerís review of Donald R, Griffinís book, Animal Minds, was titled, "A Retreat before the Canon of Parsimony" and included the following:
C. Lloyd Morganís Canon of Parsimony is not mentioned in the index but it casts a long shadow over this important book.
Barrows (1995, pp. 308, 358, and 385)
Barrowsí quotations appear in his book, Animal Behavior Desk Reference, which may be compared to a dictionary and/or encyclopedia. Relevant to the present work were the entries for "Morganís canon," "Ockhamís razor," and "law of parsimony." Given that Dewsbury was cited as a reference source within the entry for "Morganís canon," it may be pertinent to note that Dewsbury (cited here in Table 2 above and below) was one of the 13 individuals identified in the bookís Acknowledgment as having "...directly helped with this book...."
[p. 308] Morganís canon. [after quoting the canon, the entry continued] that is, one should interpret data using the most parsimonious explanation (Dewsbury, 1978, 10)....Syn. Law of parsimony, (Lloyd) Morganís canon (Dewsbury 1978, p. 10) See law: law of parsimony. Cf. Ockhamís razor. [ p. 358] Ockhamís razor. [included] Cf. Law: Law of parsimony; Morganís canon; simplicity. [p. 385] law of parsimony. [included] Cf. Morganís canon, Ockhamís razor.
Bekoff & Allen (1997, p. 326)
After quoting Zabel at al. who wrote, " One must be cautious about inferring complex cognitive processes when simpler explanations will suffice," Bekoff and Allen wrote that which is quoted below. Obviously, Bekoff and Allen considered Zabel et al.ís statement to be equivalent to Morganís canon. However, as can be seen, they (with the aid of an anonymous reviewer) tried to distinguish Morganís canon from parsimony.
The statement by Zabel et al. is a paraphrase of Morganís (1894) Canon:....It is possible that Morganís Canon which is concerned with the complexity of processes should be distinguished from parsimony which is concerned with the number of processes needed to explain a given behavior (as an anonymous reviewer noted).
Knoll (1997, p. 20)
Those with a fondness for neatly organized historical eras might say that Morganís Canon, as it is called, marks the end of the anthropomorphic strategy in psychology and the beginning of twentieth century behaviorism. [Paragraph break] However, Morganís Canon is a double-edged sword....It can cut up as well as down....if we cannot anthropomorphize the animals, we cannot anthropomorphize ourselves either.
Macphail (1998, p. 80)
What animal psychology needed, then, was...the theoretical discipline to interpret the results in as parsimonious a way as possible - a discipline crystallized by the British psychologist Conwy Lloyd Morgan (1852-1896) in his well known canon:....
Dewsbury (2000, p. 751)
In his classic textbook, Morgan (1894) outlined his famous canon that an animalís behavior should be interpreted in terms of the psychologically simplest processes consistent with the data. Morganís canon, and its related concept, parsimony, spread widely during this period.
Selected Quotations from the Sources Indicated in Table 1 to Illustrate the History of the Attempts to Correct or Draw Attention to the Misrepresentation of Morganís Canon. Sources and quotations are in bold print. Explanatory comments or supplementary information appear in non-bold print. _________________________________________________________________________________________________
Adams (1928, pp. 241-242)
Morganís canon, however, instead of being as commonly considered, a special case of the law of parsimony, is not related...and may on occasion work to exactly opposite effect....Here is Morgan trying to adapt the law of parsimony to psychology and violating it in the same breath by Ďmultiplying entitiesí making quantities of unnecessary assumptions.
Obviously, given the Adams (1928) quotation from this same article in Table 2, this is a dubious entry for Table 3.
Boring (1929, 1950)
In his "NOTES" at the end of the chapter in which Morganís canon was addressed, Boring (1929, p. 487; 1950, p. 498) wrote that which is quoted below. The principal difference between the two editions (there are other minor word changes) is that in 1950 (quoted here), the Latin phrase was added. The phrase is Boringís playful revision of what he erroneously believed to be the wording of Ockhamís razor (see the section, "Ockhamís razor" presented earlier in the present paper) and misidentified by Boring as the law of parsimony. Boring omitted "non" between "Entia" and "sunt" and substituted "propter" for "praeter," thereby reversing the intent of Ockhamís razor. While the quotation may have been intended to have a correcting influence on the interpretation of Morgan's canon, nevertheless, in view of the errors here, the anti-anthropomorphic misrepresentation of Morganís canon here, and Boring's misrepresentations as shown in Table 2, this is a highly dubious entry for Table 3.
Lloyd Morgan was justified in using it [the canon] against the tendency to anthropomorhize animals..., but conditions have changed today in comparative psychology. Cf. D. K. Adamsís criticism of the use of the law [of parsimony], The inference of mind. Psychol. Rev., 1928, 35, 235-252. Entia sunt multiplicanda propter necessitatem. [Entities are to be multiplied on account of necessity.]
Nagge (1932, p. 492-493)
Lloyd Morgan...has laid down a canon of interpretation which has come to be known to psychologists as the law of parsimony....This canon seems to have undergone a transformation in general psychological usage until it might now be tentatively expressed thus: of any possible number of explanations of an animal act the simplest possible explanation should be employed....
Nagge then explained the canon appropriately and addressed issues related to the law of parsimony. It is clear from a fuller reading of Nagge that "undergone a transformation" and "come to be known...as the law of parsimony" was seen by him as reflecting a misinterpretation of Morganís canon that should be corrected.
Newbury (1954, p. 73)
Aside from their historical inaccuracies, many current misinterpretations of Morganís Canon have sui generis failed to take advantage of possible logical developments. Without contending that Morganís methodology represents the last word, one can recognize in it some of the essentials for integrating modern introspective and comparative psychology. Whether this gain through historical continuity can be realized depends upon an accurate and significant interpretation of that methodology, including the Canon.
Before concluding the above, Newbury had cited 10 references where the canon had been misinterpreted as a version of the law of parsimony, 7 (including 5 of the parsimony-10) where it was interpreted as a doctrine of simplicity, and 4 (all among the parsimony-10) where it was related to Occamís razor. Newbury also cited other forms of misinterpretation, and he provided a detailed analysis of how Morganís canon should be interpreted.
Miller (1962, p. 214-215)
Please note Miller's second sentence.
Subsequent generations of psychologists have called this Lloyd Morganís canon and have assumed that what he must have meant was that anthropomorphism - attribution of human characteristics to gods or, as in this case animals - is unscientific. A glance into Morganís books, however, is enough to refute this assumption. Like all of his contemporaries, Morgan took it for granted that since the only psychical faculties we can know anything about directly are our own, "introspection must inevitably be the basis and foundation of all comparative psychology."3 [footnote 3 is Morgan (1894, p. 37).] Any Human introspections would necessarily be anthropomorphic; all that Morgan hoped for were a few reasonable rules for playing the game.
Gray (1963, pp. 221-222)
Gray, who provided his own analysis of what Morgan intended, identified some of the misinterpretations of Morganís canon and ridiculed them as the following examples show.
Boring, Flugel, and Skinner have referred to the Canon as a law of parsimony. Had it been such a law, surely it would have reduced Morganís entities; instead, it was compatible with their multiplication....[Paragraph break] Likewise, Thorpeís assumption that the Canon is related to Occamís razor is merely gratuitous....[Paragraph break] Watersí statements [e.g., Waters, 1939, in Table 2], not only are contrary to historical fact, but are also incorrect.
Singer (1981, p. 268)
Some workers took this principle too seriously and would not allow any interpretation of an advanced process, even if suggested by the evidence, and in 1900 Lloyd Morgan was obliged to add the following rider to his canon: ĎTo this it may be added - lest the range of the principle be misunderstood - that the canon by no means excludes the interpretation of a particular act as the outcome of higher mental processes if we already have independent evidence of their occurrence in the agent (Animal Behavior). Unfortunately, Lloyd Morganís rider has been almost totally ignored and the errors of one kind have been replaced by errors of the other.
Burghardt (1985, p. 912)
This influential statement, Lloyd Morgan's canon, is probably all that most modern psychologists know about Lloyd Morgan. It is not to be interpreted, as it commonly is, as the "simplest explanation is best." Indeed, an objection to the canon, that Morgan himself confronted is "that by adopting the principle in question we may be shutting our eyes to the simplest explanation of the phenomena" (p. 54). He then discussed examples showing that "the simplicity of an explanation is no necessary criterion of its truth" (p. 54). For example, the origin of life by creation is simpler than the indirect methods of evolution, and geological catastrophism is simpler than uniformitarianism.
Costall (1993, pp. 116-117)
Costallís entire paper is a discussion of the misrepresentation of Morganís canon, so choosing excerpts to quote was difficult. I have inserted within brackets the citations to which Costall referred with his footnotes. The two references not otherwise addressed in the present paper (viz., Richards, 1987, and Walker, 1983) have also been added to the References section for the present paper.
Later commentators have consistently represented this [Morganís canon] as an appeal to Occamís razor, a principle of parsimony; they have taken it as an outright prohibition against treating animals as anything other than mechanical automata; and they have characterized it as a rejection of anthropomorhism. [Six paragraphs later.] Morganís canon as currently misconstrued has very much the character of a myth.31 [Gray, 1963] Indeed, many of those wishing to counter the implications of this myth have themselves managed to perpetuate the myth itself. 32 [Walker, 1983] It has evidently been highly resistant to several attempts at correction.33 [Burghardt, 1985; Newbury, 1954; Singer, 1981] Indeed the two most informative recent accounts of Morganís work make no attempt to question the accepted view of the canon.34 [Boakes, 1984; Richards, 1987]
Wozniak (1993, ix-x)
After quoting examples where Skinner (1938), Griffith (1943), and Harriman (1947, and see table 2 here) associated Morganís canon closely with the law of parsimony, Wozniak wrote:
On thing is virtually certain - neither Skinner, nor Griffith, nor Harriman could ever have read Lloyd Morgan. Even if one set out deliberately to distort the meaning of Morganís canon, it would be virtually impossible to do so with greater success. Morganís canon is not a principle of parsimony,8 it was not formulated as a guide to the description of behavior, it does not dispense with mental faculties, it is not an appeal to the observable, and it is not meant to be specific to animal psychology. Even worse, it is consciously anthropomorphic and based squarely on the adequacy of the psychologistís personal introspection.
8It would be an interesting study in itself to trace the progressive distortion of Morganís views and in particular the attribution to Morgan of the principle of parsimony. Although earlier writings may also have misinterpreted Morgan in this fashion, it seems likely that Boring [Boring, E. G. (1929). A history of experimental psychology. NY: Century] was one of the more influential culprits. See especially pp. 464-465. [And Table 2 here.]
Costall, Clark, & Wozniak (1997, p. 66)
Morganís canon has been consistently overinterpreted. It was not a prohibition against the application of intentionalistic descriptions to animals, but rather an attempt by Morgan to put Ďanthropomorphismí on a more secure scientific footing (Costall, 1993).
Costall (1998, p. 18)
When Morgan realized his intentions were being misinterpreted, he added the clarification that "the Canon by no means excludes the interpretation of a particular activity in terms of the higher mental processes, if we already have independent evidence of the occurrence of these higher processes in the animal under observation (Morgan, 1903, p. 59). Nor, contrary to most accounts, was the canon, in any simple sense, an appeal to the principle of parsimony - an invitation to be economical with the truth....His serious point was that there were very good Darwinian reasons for supposing that animals should vary in the nature of their mentality. The canon was, therefore, Morganís attempt to put "anthropomorphism," the psychological approach to animals, on a secure scientific footing (Costall, 1993). [Paragraph break.] The extent to which the intentions of Morganís canon have been misinterpreted is astonishing.
Thomas (1998, p. 156)
Clearly Morganís canon was intended to be a stricture to guide the interpretation of evidence pertaining to psychological processes in animals, but the misrepresentation of that canon that occurred early...and that continues in the present...is that it was a canon of parsimony or simplicity.
The Representation of Morganís canon by Contemporary History of Psychology Textbooks.
A "Yes" entry means Morganís canon was misrepresented in that category, e.g., Benjamin (1997)
wrote, in effect, that Morganís canon was intended to be anti-anthropomorphic. A "No" entry
means Morganís canon was not misrepresented in that category, e.g., Benjafield (1996) wrote,
in effect, that Morganís canon was not intended to be anti-anthropomorphic. Blank spaces mean
that category was not addressed, this includes books that did not address Morganís canon.
Morganís Canon Advocated:
Categories: Parsimony Ockham's Simplicity Anti- Anti-
Benjafield, 1996 Yes No
Benjamin, 1997 Yes Yes
Brennan, 1998 Yes Yes
Categories: Parsimony Occamís Simplicity Anti-anthro. Anti-anec.
Goodwin, 1999 Yes Yes Yes
Hergenhahn, 2001 Yes Yes
Hothersall, 1995 Yes
Hunt, 1993 Yes
Categories: Parsimony Occamís Simplicity Anti-anthro. Anti-anec.
Leahey, 2000/2001 Yes
Schultz&Schultz, 2000 Yes No
Smith, 1997 No No
Thorne&Henley, 2001 No No No No
Viney & King, 1998 Yes Yes Yes
Categories: Parsimony Occamís Simplicity Anti-anthro. Anti-anec.
Watson/Evans, 19913 ?
1. Bolles addressed Morganís canon minimally, appropriately, but not these categories.
2. Neither Fancher nor Guthrie addressed Morganís canon.
3. Watson and Evans referred to Morgan as being anti-anecdotal, and it was suggestive but
unclear whether that was meant to contribute to the impetus for the canon (see p. 329).
Author Note: Six weeks prior to posting this article, I asked 10 scholars with credentials in history of psychology and/or comparative psychology to review the paper. Darryl Bruce, Nancy Innis, Edward Mulligan, and Robert Wozniak provided helpful criticism and suggestions for improvement, all of which were considered carefully and most of which were implemented. I am very grateful to them. Three of the remaining six replied and said that they would try to review it but had not done so by the time the article was to be posted, and three had not responded. Of course, those who reviewed the paper are blameless for any errors that may remain.