mechanistic (proximate) level and therefore, we will argue, at the ultimate
evolutionary level as well. In particular, interactions between sound waves
and the vocal-production system place significant constraints on what
sounds can be effectively generated or transmitted, thereby rendering a
large class of signals that might be theoretically possible and biologically
advantageous impossible to produce in practice. On the other hand, interactions between sound and body can in other cases provide information “by
default” without any need to invoke biological advantage or selection at all.
This is particularly true for information about body size. Finally, because
it is the dimensions of the vocal-production system that are acoustically
relevant, and not overall body dimensions, it is sometimes possible for
organisms to evade physical constraints by changing dimensions of vocal
structures independent of body size. Over the course of evolution, nature
has been ingenious in finding ways to pack more vocal tract into less body.
Thus, the communication of body size provides an ideal arena within which
to explore the interactions of physical constraints with ubiquitous selective
forces in the evolution of communication.
Body size is a critical parameter in virtually all aspects of biology. An
animal’s body mass has important implications for its physiology (SchmidtNielsen 1984), ecology (Peters 1983), fecundity (Smith-Gill and Berven
1980), and life history (Calder 1984). At the behavioral level, body size plays
a role in aggressive interactions and/or mating success (Parker 1974;
Clutton-Brock et al. 1977; Clutton-Brock and Albon 1979; Modig 1996;
Schuett 1997) Thus, the accurate perception of body size is predicted to be
adaptive for a wide variety of organisms for a number of different reasons
and should constitute a ubiquitous selective force in the evolution of communication systems. There is also a ubiquitous physical limitation on signals
in that the size of various components of the sound-production apparatus
has an important effect on the acoustic output (Fant 1960; Lieberman 1984),
with larger components producing lower frequencies. Because the size of
these production components may in many cases be related to the overall
weight or length of the animal, there is good reason to expect that some
aspects of the acoustic signal may provide cues to the size of the vocalizer.
In particular, we can predict a negative correlation between body size and
any of a variety of measures of call frequency (Morton 1977). Such acoustic
cues to body size would be internally referential (providing information
about the vocalizer itself) and direct or nonarbitrary (because the link
between large size and long wavelengths is a fact of physics).
Because of the importance of body size in animal behavior, we expect
that there will often be strong selection on perceivers to make use of available acoustic cues to body size. For the same reason, however, once
perceivers are using a particular cue, we expect selection on senders to
manipulate this cue to their own advantage (Dawkins and Krebs 1978).
There may be situations (such as when retreating from a lost aggressive
contest or luring in a timid mate) in which it would be beneficial for a sender
98
W.T. Fitch and M.D. Hauser
evolutionary level as well. In particular, interactions between sound waves
and the vocal-production system place significant constraints on what
sounds can be effectively generated or transmitted, thereby rendering a
large class of signals that might be theoretically possible and biologically
advantageous impossible to produce in practice. On the other hand, interactions between sound and body can in other cases provide information “by
default” without any need to invoke biological advantage or selection at all.
This is particularly true for information about body size. Finally, because
it is the dimensions of the vocal-production system that are acoustically
relevant, and not overall body dimensions, it is sometimes possible for
organisms to evade physical constraints by changing dimensions of vocal
structures independent of body size. Over the course of evolution, nature
has been ingenious in finding ways to pack more vocal tract into less body.
Thus, the communication of body size provides an ideal arena within which
to explore the interactions of physical constraints with ubiquitous selective
forces in the evolution of communication.
Body size is a critical parameter in virtually all aspects of biology. An
animal’s body mass has important implications for its physiology (SchmidtNielsen 1984), ecology (Peters 1983), fecundity (Smith-Gill and Berven
1980), and life history (Calder 1984). At the behavioral level, body size plays
a role in aggressive interactions and/or mating success (Parker 1974;
Clutton-Brock et al. 1977; Clutton-Brock and Albon 1979; Modig 1996;
Schuett 1997) Thus, the accurate perception of body size is predicted to be
adaptive for a wide variety of organisms for a number of different reasons
and should constitute a ubiquitous selective force in the evolution of communication systems. There is also a ubiquitous physical limitation on signals
in that the size of various components of the sound-production apparatus
has an important effect on the acoustic output (Fant 1960; Lieberman 1984),
with larger components producing lower frequencies. Because the size of
these production components may in many cases be related to the overall
weight or length of the animal, there is good reason to expect that some
aspects of the acoustic signal may provide cues to the size of the vocalizer.
In particular, we can predict a negative correlation between body size and
any of a variety of measures of call frequency (Morton 1977). Such acoustic
cues to body size would be internally referential (providing information
about the vocalizer itself) and direct or nonarbitrary (because the link
between large size and long wavelengths is a fact of physics).
Because of the importance of body size in animal behavior, we expect
that there will often be strong selection on perceivers to make use of available acoustic cues to body size. For the same reason, however, once
perceivers are using a particular cue, we expect selection on senders to
manipulate this cue to their own advantage (Dawkins and Krebs 1978).
There may be situations (such as when retreating from a lost aggressive
contest or luring in a timid mate) in which it would be beneficial for a sender
98
W.T. Fitch and M.D. Hauser