They cannot speak. Their movements are limited and clumsy. And, their sensory systems are barely functional. Evidence of habituation and associative learning can be obtained only when the most sensitive and creative behavioral testing methods are deployed. Of course, there are no signs that they reason about higher-order relations between events. Should the absence of evidence of reasoning about higher-order relations be counted as incontrovertible evidence of absence in these creatures?

This is a trick question! We might be talking about a newly hatched pigeon or we might be talking about a newborn human infant. These two organisms traverse dramatically different developmental trajectories to adulthood. As adults, pigeons fail some of the tests of higher-order relational cognition – like the forming of analogies – that humans pass. Why? Penn et al. point to the inherited information processing systems of the respective organisms; humans are born with neural systems which pigeons lack.

Penn et al.'s proposal is certainly plausible. But precisely what are these neural systems? Do these systems merely mature as the child approaches adulthood? Or must these systems be carefully cultivated by enriching experiences to fully flower? Suppose that these experiences are not a part of pigeons' usual upbringing; could providing pigeons with these experiences promote still loftier levels of cognitive achievement? Is it not reasonable to take these questions seriously before yet again proclaiming an evolved human uniqueness?

The history of comparative psychology is replete with confident proclamations of human exclusivity. Most famous and germane here is that of John Locke, who in his 1690 volume, An Essay Concerning Human Understanding, assuredly opined: “I think, I may be positive … That the power of Abstracting is not at all in [Brutes]; and that the having of general Ideas is that which puts a perfect distinction betwixt Man and Brutes; and is an Excellency which the Faculties of Brutes do by no means attain to” (Locke Reference Locke1690/1975, p. 159).

Nearly two centuries later, the faculty of abstraction was a focal point of Charles Darwin's consideration of animal and human intelligence in The Descent of Man: “[T]he greatest stress seems to be laid on the supposed entire absence in animals of the power of abstraction, or of forming general concepts” (Darwin Reference Darwin1874/1896, p. 83). Unlike Locke, Darwin left the door open to abstraction in animals. Darwin observed that:

Hence, abstract thinking might only emerge in species possessing a requisite ensemble of other, foundational cognitive skills.

Perceptively linking evolutionary and developmental considerations, Darwin anticipated and countered an obvious criticism: “It has been urged against the views here maintained that it is impossible to say at what point in the ascending scale animals become capable of abstraction, &c.; but who can say at what age this occurs in our young children? We see at least that such powers are developed in children by imperceptible degrees” (p. 84). It must be acknowledged that Darwin knew of no compelling or even suggestive evidence of abstraction in animals; his aim was to merely to remain receptive to evidence of abstraction in animals, even if rather rudimentary forms of abstraction were to be exhibited compared to the more highly advanced relational skills of human adults.

A quarter century later, C. Lloyd Morgan denied animals the ability to behave conceptually. Conceptualization, he wrote in An Introduction to Comparative Psychology, requires that we “neglect all that is variable and focus the attention on the uniform relation. [Then] we have reached a conception, and this conception is not concrete, particular, and individual, but abstract, general, and of universal application” (Morgan Reference Morgan1894/1896, p. 263). Morgan believed that only adult humans (not even children) are capable of conceptualization. He rejected the hypothesis of relational conceptualization in animals, “in no dogmatic spirit, and not in support of any preconceived theory or opinion, but because the evidence now before us is not … sufficient to justify the hypothesis” (p. 377).

What would that evidence be? Here, we hit a sticking point. Locke, Darwin, and Morgan were not at all helpful in saying just what evidence would convincingly document abstract relational behavior, particularly in nonhuman animals. Nor do Penn et al., although they do dismiss all prior claims of relational learning in nonhumans. Physical matching-to-sample, same-different discrimination, relational matching-to-sample, and both physical and functional analogy completion are all tossed aside, either because the tasks are deemed to be inadequate or because the behaviors of animals on these tasks do not meet the authors' standards. I do wish that Penn et al. had described clear behavioral tests of human and nonhuman minds that would enable investigators to see if nonverbal creatures can exhibit cognitive performances that all would agree are “truly” higher-order, relational, and systematic. Doing so would have represented a positive proposal in contrast to the rather bleak assessment of animal cognition that Penn et al. now provide.

This negative assessment prompts Penn et al. to assert that Darwin was mistaken when he proposed that intellectual differences between animals and humans are quantitative and not qualitative. I trust that the earlier quotations document that Darwin held a fuller and more nuanced position concerning species differences in intelligence than is suggested by his oft-cited “continuity” claim.

Finally, Penn et al. should clarify what appears to be a striking anomaly in their logic. After strenuously arguing that human and animal minds differ in kind, Penn et al. curiously suggest that human and nonhuman minds differ in the degree to which they approximate the relational capabilities of a physical symbol system. Can they have it both ways?