In this thought-provoking article, Ramsey et al. develop a rationale for a practical method to study behavioral novelty under field conditions. The virtue of this operational definition is that it lends itself to field studies with relatively narrow time constraints. As happens in the best of proposals, virtues and weaknesses arise from the same source, in this case the constraints of relatively short-term field observations. The authors clearly point out this inherent weakness, but choose instead to devote discussion to details of method and positive results – as well they should. A strength of this article is that in the course of presenting an ingenious operational definition and its implications, Ramsey et al. draw attention to central issues of innovation and incipient signs of culture in nonhuman animals.

Incipient culture. Ramsey et al. discuss lateral spread of new behaviors from individual to individual within a social group much more than vertical transmission from generation to generation. Yet, transmission across generations, especially from parent to offspring, seems closer to the concept of human culture than lateral transmission. In human cases, lateral spread from individual to individual within a single generation could be fad or fashion rather than culture. Boesch-Achermann and Boesch (Reference Boesch-Achermann and Boesch1993), for example, observed wild chimpanzees using tools to crack nuts. Nut cracking with tools must have originated as a novel behavior sometime in the history of this social group of chimpanzees. Implications for a kind of culture of nut cracking depend, however, on transmission across generations, particularly on observations of mothers teaching infants (Boesch Reference Boesch1991).

Long-term field studies tend to require monumental effort and time, but can also yield monumental results (Goodall Reference Goodall1986; Watanabe Reference Watanabe, Gardner, Gardner, Chiarelli and Plooij1994). However, cross-generational transmission appears in laboratory experiments. Previde and Poli (Reference Previde and Poli1996), for example, taught a novel method of food retrieval to female hamsters. Pups observed their mothers and later incorporated the novel method into their – second generation – food retrieval.

Fostering culture. Cross-fostering – parents of one genetic stock rearing the young of a different genetic stock – is a traditional tool of behavioral biology for studying the interaction between genetic endowment and developmental environment (Immelmann & Beer Reference Immelmann and Beer1989; Scott Reference Scott1958; Stamps Reference Stamps2003). Members of fostered species have adopted species-specific behavior of fostering parents, for example, migratory habits (Harris Reference Harris1970) and flight and feeding habits (Rowley & Chapman Reference Rowley and Chapman1986). In the context of the present discussion, behaviors adopted by cross-fosterlings are clearly novel to the fostered species. Their adoption across generations looks very much like culture.

Sign language studies of cross-fostered chimpanzees (Gardner et al. Reference Gardner, Gardner and Van Cantfort1989) take the method of cross-fostering farther because chimpanzees resemble humans so closely, particularly in their comparably long childhood (Gardner & Gardner Reference Gardner, Gardner, Gardner, Gardner and Van Cantfort1989, pp. 4–5). Cross-fostering is very different from rearing a chimpanzee in a conventional laboratory staffed by human caretakers. Cross-fostering is also very different from keeping a chimpanzee in a home as a pet. Many people keep pets in their homes. They may treat their pets very well, and they may love them dearly, but they hardly treat them like children. Providing a nearly human infant environment all day every day for years on end is a daunting laboratory challenge (Kellogg Reference Kellogg1968).

All aspects of intellectual growth are intimately related. For young chimpanzees, no less than for human children, familiarity with simple tools such as keys, devices such as lights, articles of clothing such as shoes, are intimately involved in learning signs or words for keys, lights, shoes, opening, entering, lighting, and lacing. The human-simulated homes of chimpanzees, Washoe, Moja, Pili, Tatu, and Dar were well-stocked with human objects and activities, and cross-fosterlings had free access to them, or at least as much access as young human children usually have. They ate human style food at a table, with cups, forks, and spoons. They helped to clear the table and wash the dishes after a meal. They used human toilets (in their own quarters and elsewhere), wiped themselves and flushed the toilet, and even asked to go to the potty during boring lessons and chores (Gardner & Gardner Reference Gardner, Gardner, Gardner, Gardner and Van Cantfort1989).

The daily language of this infant world was American Sign Language (ASL), the naturally occurring language of deaf communities in North America. English, the language of earlier studies, demands a vocal apparatus and vocal habits that seem to be beyond chimpanzees. Without conversational give-and-take in a common language, cross-fostering conditions could hardly simulate the environment of a human infant. Whenever a cross-fosterling was present all verbal communication was in ASL. Total immersion in human language as well as human culture is essential for rigorous, human cross-fostering.

A naturally occurring human language acquired in a human cultural context permitted human comparisons. Size of vocabulary, appropriate use of sentence constituents, number of utterances, proportion of phrases, and inflection all grew robustly throughout five years of cross-fostering; robustly, but more slowly than in human children. Growth was patterned growth and patterns were consistent across chimpanzees. Wherever there are comparable measurements, patterns of growth for cross-fostered chimpanzees paralleled in detail characteristic patterns reported for human infants (B. Gardner & Gardner Reference Gardner and Gardner1998; R. Gardner & Gardner Reference Gardner and Gardner1998). All of ASL was novel to chimpanzees, of course.

Pragmatic devices that characterize human conversation also appeared in conversations of cross-fostered chimpanzees (Bodamer & Gardner Reference Bodamer and Gardner2002; Chalcraft & Gardner Reference Chalcraft and Gardner2005; Drumm et al. Reference Drumm, Gardner and Gardner1986; Jensvold & Gardner Reference Jensvold and Gardner2000; Shaw Reference Shaw2000). Conversational pragmatics of intensity, agreement, and turn-taking seem more cultural than Chomsky's innate deep structures. Cross-fostered chimpanzees converse with each other in the absence of any human presence with only video cameras to make records (Fouts & Fouts Reference Fouts, Fouts, Gardner, Gardner and Van Cantfort1989). Project Loulis adds a further cultural dimension. The infant, Loulis, adopted by Washoe when he was about a year old, learned more than 50 signs of ASL that he could only have learned from other chimpanzees (Fouts et al. Reference Fouts, Fouts, Van Cantfort, Gardner, Gardner and Van Cantfort1989). A culturally relevant laboratory succeeded where Terrace et al. (Reference Terrace, Pettito, Sanders and Bever1979) failed with rigorous Skinnerian reinforcement. Human-like development depends on human cultural context. Ramsey et al. help to place a traditional biological method in philosophical and anthropological context.