Kline's attempt to bring together the various disciplines that study teaching behavior, and their disparate perspectives, is commendable. The target article's central theme – that a synthesis of comparative and cross-cultural research on the topic could move our understanding significantly forward – is well taken. Although most researchers would agree with the definition of teaching “as behavior evolved to facilitate learning in others” (target article, sect. 1, para. 5), vigorous debate has reigned beyond this most basic point. Fueling much of the controversy has been a proliferation of terminology without enough attention to consistency, as well as a penchant to disregard the complementarity of different levels of explanation (sensu Tinbergen Reference Tinbergen1963). Psychologists have tended to exclusively focus on the proximate cognitive mechanisms that may serve to support teaching; this is what Kline terms the “mentalistic” approach. Kline's “functionalistic” approach, which is taken by many evolutionary anthropologists and biologists, is an ultimate-level perspective. Unfortunately, the functionalistic framework has not encouraged a nuanced exploration of either the potential variation in teaching behavior that may be exhibited across nonhuman species or of the cognitive processes that may be involved (Byrne & Rapaport Reference Byrne and Rapaport2011; Rapaport & Byrne Reference Rapaport and Byrne2012). Instead, much of the emphasis of this approach has centered on cataloguing species that fit Caro and Hauser's (Reference Caro and Hauser1992) narrow operational definition (Thornton & Raihani Reference Thornton and Raihani2008; Reference Thornton and Raihani2010).
Such single-minded attention has left uncontested the potentially erroneous argument that teaching in nonhumans is merely analogous to teaching in human. Consequently, the claim that comparative study cannot not inform us as to the evolutionary sequence that led to intentional teaching in humans (Csibra & Gergely Reference Csibra and Gergely2011) remains untested. This humans-only perspective may well be premature. Therefore, I agree with the need for synthesis but would argue that there is little justification for introducing new terms for these various approaches, given the already vast terminology that plagues the field and the enduring utility of Tinbergen's (Reference Tinbergen1963) levels-of-explanation paradigm.
More importantly, I question the justification for expanding the characterization of teaching to include situations in which the knowledgeable partner does not actively promote learning but simply tolerates close and intrusive observation by a naïve individual (“teaching by social tolerance” in Kline's parlance). Young nonhuman primates often share feeding sites with and forage in close proximity to adult conspecifics, especially close relatives, who also are feeding and foraging. True, tolerance may be costly when the young interfere with adult feeding efficiency, but when adults do not actively assist immatures, any food-related information they gain from these activities is realized through efforts initiated by the young (reviewed in Rapaport & Brown Reference Rapaport and Brown2008). When the presence of another individual leads an observer behave in a similar way or to focus its attention on a location or object, without active assistance from the knowledgeable partner, the social learning that occurs is thought to be inadvertent on the part of the demonstrator. Only when the knowledgeable individual alters its behavior in order to demonstrate relevant environmental cues or behavior has the interaction been considered to be “teaching” (Hoppitt et al. Reference Hoppitt, Brown, Kendal, Rendell, Thornton, Webster and Laland2008).
Alternatively, Kline proposes that teaching by social tolerance can be recognized when “the degree of tolerance … is greater than species-typical tolerance toward other conspecifics” or by “heightened social tolerance in situations … when the potential gains in learning for the young are especially high” (sect. 3.1.1). The problem is, adult primates are noticeably tolerant of closely related young in many different situations. When a mother macaque allows her juvenile to use her back as a springboard as she rests in the midday heat or when her infant hangs from her arm as she attempts to groom another adult, is the mother engaged in teaching? One could argue that immature primates learn adaptive lessons in social etiquette or acquire physical training through these activities (Bekoff & Byers Reference Bekoff and Byers1998; Byers & Walker Reference Byers and Walker1995), and that adults suffer energetic or opportunity costs from intrusions by the young, but these situations appear to be cases in which immatures are exhibiting an adaptive propensity to take advantage of adult tolerance in order to facilitate individual learning. Thus, the learning that occurs is likely to be less socially based than that which may occur during co-feeding and co-foraging, and yet, the behavior of the adult – that is, tolerance – is not measurably different.
Likewise, “teaching by opportunity provisioning” appears to not always incorporate active involvement by the putative demonstrator. This category comprises situations in which the knowledgeable partner allows access to information that is otherwise too difficult or too dangerous for the naïve pupil to acquire by individual means. Yet the human example that Kline provides – that of a music teacher who “plays music for his students without modification” (sect. 4.2.2.) – evinces no behavioral adjustment on the part of the teacher. Kline's overly broad definitions open the door to a host of social-learning situations in which the demonstrator simply goes about his or her normal activities, inclusion of which may only muddy the waters, rather than contribute to a greater understanding of what it means to teach. Hoppitt and colleagues (Reference Hoppitt, Brown, Kendal, Rendell, Thornton, Webster and Laland2008) have provided a scheme that is similar to Kline's, which recognizes that the social learning processes from the perspective of the pupil may not substantially differ in situations that involve or do not involve active teaching by a knowledgeable partner. However, Hoppitt et al.'s scheme more clearly defines teaching as incorporation of active demonstration. On the other hand, the present article offers the valuable addition of a category, direct active teaching, which thus far has been convincingly reported only in humans. The most valuable contribution, however, may be Kline's emphasis that social learning in general, and teaching specifically, form overlapping continua in which multiple strategies may be employed at any given time. This is one of the characteristics of teaching behavior that creates difficulties for investigators, but makes the behavior so endlessly interesting.
You have
Access
Kline's attempt to bring together the various disciplines that study teaching behavior, and their disparate perspectives, is commendable. The target article's central theme – that a synthesis of comparative and cross-cultural research on the topic could move our understanding significantly forward – is well taken. Although most researchers would agree with the definition of teaching “as behavior evolved to facilitate learning in others” (target article, sect. 1, para. 5), vigorous debate has reigned beyond this most basic point. Fueling much of the controversy has been a proliferation of terminology without enough attention to consistency, as well as a penchant to disregard the complementarity of different levels of explanation (sensu Tinbergen Reference Tinbergen1963). Psychologists have tended to exclusively focus on the proximate cognitive mechanisms that may serve to support teaching; this is what Kline terms the “mentalistic” approach. Kline's “functionalistic” approach, which is taken by many evolutionary anthropologists and biologists, is an ultimate-level perspective. Unfortunately, the functionalistic framework has not encouraged a nuanced exploration of either the potential variation in teaching behavior that may be exhibited across nonhuman species or of the cognitive processes that may be involved (Byrne & Rapaport Reference Byrne and Rapaport2011; Rapaport & Byrne Reference Rapaport and Byrne2012). Instead, much of the emphasis of this approach has centered on cataloguing species that fit Caro and Hauser's (Reference Caro and Hauser1992) narrow operational definition (Thornton & Raihani Reference Thornton and Raihani2008; Reference Thornton and Raihani2010).
Such single-minded attention has left uncontested the potentially erroneous argument that teaching in nonhumans is merely analogous to teaching in human. Consequently, the claim that comparative study cannot not inform us as to the evolutionary sequence that led to intentional teaching in humans (Csibra & Gergely Reference Csibra and Gergely2011) remains untested. This humans-only perspective may well be premature. Therefore, I agree with the need for synthesis but would argue that there is little justification for introducing new terms for these various approaches, given the already vast terminology that plagues the field and the enduring utility of Tinbergen's (Reference Tinbergen1963) levels-of-explanation paradigm.
More importantly, I question the justification for expanding the characterization of teaching to include situations in which the knowledgeable partner does not actively promote learning but simply tolerates close and intrusive observation by a naïve individual (“teaching by social tolerance” in Kline's parlance). Young nonhuman primates often share feeding sites with and forage in close proximity to adult conspecifics, especially close relatives, who also are feeding and foraging. True, tolerance may be costly when the young interfere with adult feeding efficiency, but when adults do not actively assist immatures, any food-related information they gain from these activities is realized through efforts initiated by the young (reviewed in Rapaport & Brown Reference Rapaport and Brown2008). When the presence of another individual leads an observer behave in a similar way or to focus its attention on a location or object, without active assistance from the knowledgeable partner, the social learning that occurs is thought to be inadvertent on the part of the demonstrator. Only when the knowledgeable individual alters its behavior in order to demonstrate relevant environmental cues or behavior has the interaction been considered to be “teaching” (Hoppitt et al. Reference Hoppitt, Brown, Kendal, Rendell, Thornton, Webster and Laland2008).
Alternatively, Kline proposes that teaching by social tolerance can be recognized when “the degree of tolerance … is greater than species-typical tolerance toward other conspecifics” or by “heightened social tolerance in situations … when the potential gains in learning for the young are especially high” (sect. 3.1.1). The problem is, adult primates are noticeably tolerant of closely related young in many different situations. When a mother macaque allows her juvenile to use her back as a springboard as she rests in the midday heat or when her infant hangs from her arm as she attempts to groom another adult, is the mother engaged in teaching? One could argue that immature primates learn adaptive lessons in social etiquette or acquire physical training through these activities (Bekoff & Byers Reference Bekoff and Byers1998; Byers & Walker Reference Byers and Walker1995), and that adults suffer energetic or opportunity costs from intrusions by the young, but these situations appear to be cases in which immatures are exhibiting an adaptive propensity to take advantage of adult tolerance in order to facilitate individual learning. Thus, the learning that occurs is likely to be less socially based than that which may occur during co-feeding and co-foraging, and yet, the behavior of the adult – that is, tolerance – is not measurably different.
Likewise, “teaching by opportunity provisioning” appears to not always incorporate active involvement by the putative demonstrator. This category comprises situations in which the knowledgeable partner allows access to information that is otherwise too difficult or too dangerous for the naïve pupil to acquire by individual means. Yet the human example that Kline provides – that of a music teacher who “plays music for his students without modification” (sect. 4.2.2.) – evinces no behavioral adjustment on the part of the teacher. Kline's overly broad definitions open the door to a host of social-learning situations in which the demonstrator simply goes about his or her normal activities, inclusion of which may only muddy the waters, rather than contribute to a greater understanding of what it means to teach. Hoppitt and colleagues (Reference Hoppitt, Brown, Kendal, Rendell, Thornton, Webster and Laland2008) have provided a scheme that is similar to Kline's, which recognizes that the social learning processes from the perspective of the pupil may not substantially differ in situations that involve or do not involve active teaching by a knowledgeable partner. However, Hoppitt et al.'s scheme more clearly defines teaching as incorporation of active demonstration. On the other hand, the present article offers the valuable addition of a category, direct active teaching, which thus far has been convincingly reported only in humans. The most valuable contribution, however, may be Kline's emphasis that social learning in general, and teaching specifically, form overlapping continua in which multiple strategies may be employed at any given time. This is one of the characteristics of teaching behavior that creates difficulties for investigators, but makes the behavior so endlessly interesting.