point what may now be unconscious was consolidated through the conscious
and through learning.
Scientists generally agree that for humans, language and communication
involve a high degree of teaching and learning. Yet there is a reluctance among
ethologists to accept that nonhuman animals transfer information through com-
munication. Dawkins and Krebs (1978, p. 308), quoted in Seyfarth et al. (2010),
claim that “it is probably better to abandon the concept of information transfer
altogether,” apparently agreeing with Owings and Morton (1997, 1998) that to
do so is both anthropomorphic and inaccurate. Seyfarth et al. conclude that ani-
mal calls do contain information that affects the receiver’s responses and can
illicit more than one response and that “animal signals encode a surprisingly
rich amount of information” (p. 7). They arrive at these conclusions without
discussing any of the findings on elephant communication and with no recogni-
tion that much of that communication is in the infrasonic range, not audible to
humans, and in the broader sense poorly observed and understood.
Caro and Hauser (1992), together with Thornton and Raihani (2008), be-
lieve that teaching must be explicit, although not dependent upon, one teacher
and one learner and must result in the pupil acquiring new skills or competence.
While acknowledging the role of mothers and social learning, they do not con-
sider this to be a true form of teaching. Surprisingly, they include no discussion
of play as a form of teaching or learning. Teaching, however, should result in
learning a new skill or acquiring new knowledge. Teaching should result in
changes in behavior and the acquisition of new skills.
Much of elephant learning must be through observing group members, with
individuals spending an entire lifetime within a single group continuously fol-
lowing older more knowledgeable relatives. The nearly 2 years that a calf nurses
followed by 8–10 years of close contact with the mother and attention paid
Elephant Sense and Sensibility. http://dx.doi.org/10.1016/B978-0-12-802217-7.00012-0
Copyright © 2015 Elsevier Inc. All rights reserved.
91
92 Elephant Sense and Sensibility
by allomothers to the calf provide an extended period of learning at a critical
time of development for the calf. Almost uniquely among mammals, female
elephants remain within a closely knit family unit for the rest of their lives.
Behavior learned within this group consists not only of the crucial knowledge
needed for survival but equally important knowledge involving social skills and
interactions with their family and a much wider population that may number
well over 100 individuals.
In contrast, males leave the herd at puberty and show distinctly different
upbringing when compared to females. For example, crop-raiding is a high-
risk, high-reward exercise that is carried out mostly by male elephants. Energy
requirements of males rise at the age of first reproduction (25–30 years old).
At their reproductive peak (45–50 years), males’ nutritional demands are at the
highest level. Moss and her colleagues (Chiyo et al., 2011) have shown that
males at their prime take more risks than younger males or females. Females in
the Amboseli herds, although exposed to crops, do not take the risks entailed in
crop-raiding. Younger males, however, join with experienced older males, sug-
gesting that they have found that social learning is better than solitary learning,
especially when the cost of exploratory learning may be high.
Failure to observe teaching in the case of elephants may be a function of
the lack of opportunity to observe such a process rather than the conclusion
that teaching among elephants does not exist. A more reasonable conclusion
may be that explicit teaching is not a frequent function in elephant society and
as such rarely observed. A potentially explicit case of teaching by an elephant
mother of its calf was observed in a dry riverbed in the Kruger National Park by
the author (Figure 12.1). The mature female, possibly the matriarch of a small herd, had successfully dug for water in the sand of a dry riverbed. Her calf, perhaps 4–5 years old, was obviously interested in the water at the bottom of a 1 m
(3 ft) deep hole. However, it neither knew how to drink using its trunk nor how
to reach the water at the bottom of the hole. The mother took up a quantity of
water in her trunk. She wound her trunk around that of her calf such that the tip
of her trunk was immediately below the tip of her calf’s trunk. She then allowed
the water in her trunk to flow out such that it welled up into the tip of the calf’s
trunk. It seemed obvious to the observer that she was trying to get the calf to
take up the water in its trunk. The calf spluttered and struggled but failed to get
the idea. The mother tried repeatedly to get the calf to respond but after about
four or five tries gave up, took a trunk full of water and pumped it directly into
the calf’s mouth. From the sequence of actions taken by the mother it seemed
that she had the clear intent of getting the calf to use its trunk as a means to
drink. She clearly knew that the calf had to take up the water in the lower end
of its trunk and transfer this water to its mouth. She did everything she knew to
show the calf how to do this. When this failed she knew that the transfer process
from the trunk to the mouth by the calf was not going to work and that she had
no option but to execute this part of the operation herself. Teaching may thus be
an uncertain process that needs to be repeated many times before the lesson is
Learning and Teaching Chapter | 12 93
FIGURE 12.1 Mother repeatedly tried to get the calf to take water she had drawn from a hole dug in the river bed too deep for the calf to reach. (Pen and ink drawing by the author.)
learned. It is possible that elephants perform such repetitive instruction teaching
young by multiple subtle demonstrations.
Earlier we described a lone female African forest elephant who seldom if
ever vocalized except when apparently hearing distant low-frequency sounds
generated by the wake vortices from a 747 Boeing aircraft taking off from the
Lahore airport. While there is no evidence that this elephant mimics the sounds
she heard, more recent observations have documented that elephants can learn
to imitate the sounds of truck engines (Poole et al., 2006). Stoeger et al. (2012)
has documented that male African elephants living with two female Asian ele-
phants have learned to mimic the chirping sounds made by the Asian elephants.
Stoeger also documents an Asian elephant mimicking human speech.
Vocal learning not only means that elephants are aware of sounds made by
other animals and nonanimal sources but that they have the potential of devel-
oping an open communication system that potentially could include humans.
Byrne and Bates (2009, p. 72) suggest that there seems to be some teaching
94 Elephant Sense and Sensibility
behavior inherent in the attention paid by older females to a young female when
she first comes into estrous.
de Waal ( The Bonoko and the Atheist, 2013, pp. 114–115) describes an ex-
perime
nt conducted at the Washington National Zoo by Preston Foerder and
Deana Reiss. Kandula, a young elephant bull, was presented with bunches of
fruit being suspended out of reach from the roof of the enclosure. Several ob-
jects, including sticks, a sturdy box, and several thick cutting boards, were first
distributed around the enclosure. Kandula ignored the sticks but after a while
moved the box with his foot in a straight line until it was under a bunch of fruit.
He then stood with his front feet on the box, reaching up to the fruit with his
trunk. Foerder and Reiss then moved the box outside and out of sight from the
enclosure. Without hesitation, Kandula retrieved the box, from apparently con-
siderable distance, bringing it once again to a position below a bunch of fruit.
They then further complicated the experiment by removing the box entirely,
leaving the sticks and wooden boards. Again Kandula ignored the sticks, picked
up the wooden boards, and stacked them on top of each other to serve as a plat-
form to reach the fruit.
While elephants commonly use their trunks to reach high into trees for fruit
and other edible plant material, and will make use of terrain, roots, and other
small elevated surfaces to add to their reach, they have not been observed in the
wild to move objects to stand on (Figures 12.2 and 12.3). In the Washington Zoo case, Kandula showed not only deductive reasoning but a learning process
that he could recall, repeat, and substitute even when the means (tools) were no
longer visible or available. While this experiment was conducted under artificial
conditions, it had elements based on natural conditions that utilized both the
FIGURE 12.2 Foreign objects are readily recognized for what they are: a potential source of clean water if broken into, which they often are.
Learning and Teaching Chapter | 12 95
FIGURE 12.3 Reaching for objects perhaps as high as 8 m (25 ft) above the ground. (Pen and ink drawing by the author.)
neural and physical assets of the elephant. Being able to either observe behavior
in the natural world or effectively transfer natural simulations to artificial envi-
ronments is essential in trying to assess the intelligence of animals.
Sitting behind a desk is a dangerous place from which to view the world
(John le Carré quote; http://www.goodreads.com).
Chapter 13
The Sensory Environment
of Elephants
Sensory signals are transmitted in multiple ways, consciously and uncon-
sciously, by individuals and through group responses. Individuals transmit
signals through the five sensory systems. Conscious signals are frequently ob-
vious. My nephew, Richard Garstang, and I demonstrated the latter when be-
ing harassed by a young, fairly aggressive bull elephant in the Kruger National
Park. The young male had repeatedly mock-charged our Volkswagen “Combie.”
After about the third episode we waited for another frontal threat, then as the
youngster came toward us we simultaneously swung open the front doors of
the Combie in a realistic imitation of the flaring of the ears at an adversary. The
feisty youngster skidded to a halt, turned, and abandoned his game in immediate
response to a well-understood threat from a larger opponent.
Unconscious signals are less obvious, more subtle, and not under the con-
scious control of the animal. Pheromone emitted signaling fear or aggression
are not controlled. Other subtle signals in tone and body language may not be
consciously generated (Figures 13.1–13.4).
Collective signals may be generated by groups and large assemblies.
Humans, for example, believe that in sports events crowd energy can be trans-
ferred to players. This may well be achieved in a number of ways by sound,
smell, and collective body language. These responses have evolved over evo-
lutionary time, particularly in social and herd animals, and may be far more
prevalent in animals than in humans. Ecologists and behavioral biologists have
focused on conspecific communication and some interspecies communication
but paid scant attention to overall environmental acoustic, olfactory, visual, tac-
tile, and gustatory input.
Of the sensory input, the auditory component may be the most transform-
ing and may be far more pervasive than has been previously considered. The
sensory surroundings of an elephant consist initially of its mother’s womb and
ultimately of its total environment. Detectable initially by touch and smell and
subsequently by sound, taste, and sight, it ultimately involves all five senses,
which serve to identify the elephant’s place in its world.
Elephant Sense and Sensibility. http://dx.doi.org/10.1016/B978-0-12-802217-7.00013-2
Copyright © 2015 Elsevier Inc. All rights reserved.
97
98 Elephant Sense and Sensibility
FIGURE 13.1 Awareness and tolerance of others is exhibited by clear body language from mild curiosity to explicit warning signs.
FIGURE 13.2 Awareness and mild curiosity with no threat.
The Sensory Environment of Elephants Chapter | 13 99
FIGURE 13.3 Awareness with some fear and potential flight.
FIGURE 13.4 Awareness coupled with mild flight.
The sensory systems of mammals (auditory, olfactory, visual, tactile, and
gustatory) provide parallel and continuous input from the environment to the
brain. Neural systems must function to interpret the signals and assign meaning
to the integrated input. Interpretation must trigger memory and response to a
of responses upon which survival may have depended.
100 Elephant Sense and Sensibility
The acoustic fields in the natural environment may represent signals that
travel over the longest distances and that constitute the basis for mammals such
as elephants to recognize spatial and other characteristics of their surroundings.
There is ample evidence that suggests that elephants can travel over large
distances (hundreds of kilometers) to reach specific locations at specific times.
There is further evidence that animals who have been translocated over similar
long distances (in closed vehicles) are subsequently able to return on foot to
their place of origin. Orphan elephants at the David Shepherd Wildlife Trust
are taken by road in closed vehicles after as much as 10 years in the Trust to
be released in the Tsavo National Park approximately 150 km (93 mile) away.
Female released elephants have successfully integrated into the wild herds of
Tsavo, some of them ultimately producing calves of their own. In a number of
instances mothers have brought these calves over a distance of at least 150 km
(93 mile) back to the Trust. Quite apart from any attempt to understand the mo-
tivation of the mother to bring their calf back to the Trust, no plausible explana-
tion can be offered to explain how the mother knew where the Trust was relative
to Tsavo and how she was able to get there. Seeking a rational explanation one
is forced to call upon more than one sophisticated use of navigation, memory,
and neural competence.
In the case of Lawrence Anthony’s death, described in Chapter 7, the pos-
tulated response of the herds is to t
he behavior or, perhaps, more importantly is
to the changed behavior of another species. Not only might the elephants have
detected this change in behavior but they may have correctly interpreted it as
displaying grief.
It is likely that individual elephants have a total neural memory of the
sensory environment of their home territory. A significant part of this neural
memory consists of auditory sounds stemming from other than conspecifics
including humans. Abiotic sounds add to the input. This wide range of sounds
includes frequencies and levels that are inaudible to humans. As such, these
sounds have been poorly recorded and inadequately studied. While the acoustic
environment may extend over the greatest area, it remains a part of the total
sensory environment that defines the animal’s home territory. This total sensory
environment is made up of all of the sounds, smells, sights, and tactile and taste
sensations that have been recorded and consolidated within the animal’s brain
over the time spent in its territory. The sum total of this sensory input serves to
provide the basis for the animal not only to instantaneously recognize its home
territory but to be conscious of any changes or inconsistencies in the environ-
ment of that territory.
The embedded memory consisting of integrated multiple sensory inputs
serves as a basis to recognize familiar territory such as the component parts of
the home range. It also serves as a neural framework against which changes or
unusual circumstances can be recognized.
Recognizing that elephants have such capabilities presents the challenge of
interpretation by humans of such a complex synthesis of input but will also have
The Sensory Environment of Elephants Chapter | 13 101
important implications to management and conservation. Inherent in the propo-
sition is the concept that the neural connection to a given environment, which is
far more complex than simply a connection to the biota of that environment, is
fundamental to these animals. Changes such as translocations represent in these
terms fundamental dislocation.
Chapter 14