2.3 Organisms and Function

К оглавлению1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 
34 35 36 37 38 39 40 41 

In the last two sections of this chapter, I have argued for a nomological

form of metaphysical emergence, in terms of HOV, and I began to argue

for a representational form of epistemological emergence, in terms of an

as-if realism. In the fi nal section of this chapter, I will continue to argue

for both forms of emergence by an investigation of function (also see Arp,

2006b). That organisms have functions and that we must describe—or, at

least, as-if describe—the actual functioning of those organisms are central

to my metaphysical and epistemological versions of emergence. Further,

it is essential to my project that I explain and defend a description of functions

because my hypothesis concerning scenario visualization depends

upon certain functional mechanisms of the mind having evolved to solve

specifi c problems encountered in various Pleistocene environments. As a

biologist (of sorts), the realist in me says that organisms really do have

functions. As a philosopher of biology, the antirealist in me realizes that no

one can know, for sure, if organisms really do have functions, principally

because of the veil of perceptions that mediates between the mind and

world (if there is a world). However, we still must proceed to describe

organisms as if they have functions because (1) such a description is useful

and (2) it would be miraculous if our descriptions did not match up with

what we take to be the real functions of organisms as we perceive them.

Organisms, and the subsystems and processes that comprise them, tend

to operate in certain ways on a regular basis. This operational regularity

not only aids biologists in identifying certain traits but enables biology to

be considered an autonomous science with its own domain of laws (also

see the sections on function in Arp & Ayala, 2008; Arp & Rosenberg, 2008;

Terzis & Arp, 2008). Genetic information is communicated from parent

organism to offspring, mitochondria convert sugar to ATP, the heart pumps

blood, the medulla controls breathing, the complex interactions of the

subsystems and processes of multicellular organisms tend toward the maintenance

of a dynamic equilibrium (homeostasis)—all of this happens in

fairly predictable and reliable ways. Another way of describing the tendency

for processes and subsystems of organisms to operate in predictable

and reliable ways is in terms of the organism’s functions.

In the biological realm, a complete explanation of a trait seems to

include an explanation in terms of function. It is natural to ask of some

trait, What is its function? or What purpose in the organism does the particular

trait serve? or What is the goal of its activity? Thus, for example,

to explain the existence of the human heart as merely a red mass of tissue located in the chest cavity of the torso that pulsates and reverberates

according to the natural laws of physics and chemistry strikes us as an

incomplete explanation. However, as has been argued quite convincingly

by Mayr (1996), Walters (1998), Lennox (1993), Sober (1993), Gould (1977),

Sterelny (2001), Dawkins (1986), and others, biology has its own legitimate

scientifi c principles and terminology (cf. Machamer, Darden, & Craver,

2000; Tabery, 2004; Glennan, 1996). In line with a biological explanation,

it seems we can talk about the heart’s function or purpose in the human

body as an organ that pumps blood. An important part of explaining the

heart is to say that it exists so as to pump blood, or that pumping blood

is the heart’s operation or function.

Beginning in the 1960s with Nagel (1961) and Hempel (1965), philosophers

of science and biology have not only proposed a variety of defi nitions

for the term function but have attempted to specify the conditions

where it may be appropriate to predicate functions of natural and artifi cial

things, and to appeal to these functions in explanations (see Cummins,

1975, 2002; Wright, 1973, 1976; Mayr, 1982, 1988; Millikan, 1984, 1989,

2002; Buller, 1999; Ayala, 1998; Mahner & Bunge, 2001; Neander, 1991,

1999; Hardcastle, 1999; Perlman, 2004). Some thinkers, like Dennett (1987,

1995) and the Churchlands (1986, 1989), offer a promissory note that talk

of functions in biology will be eliminated altogether once a more accurate,

value-free system of terms is invented to describe the operational traits of

organisms. Searle (1990b, p. 414) seems to hold the similar view that functions

are solely the product of intentional and intensional minds, and that

functions are “never intrinsic but are always observer relative.” And Ruse

has affi rmed (1971, 1973) and reaffi rmed (2003) a position similar to that

of Dennett, the Churchlands, and Searle, namely, there is nothing like an

intrinsic or extrinsic ontological purpose or design to be found in nature.

Talk of functions is just that—talk of functions.

Yet Ruse and other thinkers admit something about the biological sciences

that Kant (1987) had admitted in his third Critique; with respect to

organisms, it is useful to think as if these entities have traits and processes

that function in goal-directed ways. Thinkers cannot seem to get around

Trivers’ (1985, p. 5) claim that “even the humblest creature, say, a virus,

appears organized to do something; it acts as if [italics mine] it is trying to

achieve some purpose” or Arnhart’s (1998, p. 245) observation that

“although the evolutionary process does not serve goals, the organisms

emerging from that process do. Darwin’s biology does not deny—rather,

it reaffi rms—the immanent teleology displayed in the striving of each

living being to fulfi ll its specifi c ends. . . . Reproduction, growth, feeding, healing, courtship, parental care for the young—these and many other

activities of organisms are goal-directed” (also see Bogdan, 1994; Stout,

1996; FitzPatrick, 2000; Sterelny & Griffi ths, 1999; Arp, 1998, 1999, 2002,

2006b).

In the last section, I argued for an as-if realism. This approach acknowledges

that we could never know the nature of anything outside of our

own perceptions for certain. At the same time, as-if realism affi rms that

we must act as if there is a reality out there to be understood and

described; it would be miraculous if there were not a real world out

there to which our perceptions and ideas correspond, given the fecundity

of our scientifi c, philosophic, and other research endeavors. With this

as-if realism in mind, we can still attempt to describe the functions of

the traits and processes of organisms. Thus, I am in agreement with Ruse

and other thinkers who do not attempt to explain away descriptions of

the functions of traits. The as-if realism I endorse works together nicely

with this as-if description of the functioning of biological traits—the

as-if realist in me acknowledges that organisms act as if they are organized

so as to function in specifi c ways. In fact, as noted already, it is

essential to my project that I lay out and defend a description of functions

because, as will be shown in later chapters, my hypothesis concerning

scenario visualization depends upon certain functional mechanisms

of the mind having evolved to solve specifi c problems encountered in

various Pleistocene environments.

There are several views concerning the appropriate defi nition of function

for biological matters, and Perlman (2004) has done a fi ne job of laying

out all of these views as envisioned in the history of Western philosophy.

The two views of function with respect to living things that, in my estimation,

have the most credibility are Cummins’ (1975, 2002) organizational

account and the Griffi ths/Godfrey-Smith modern history account (Griffi ths,

1992, 1993, 1996; Godfrey-Smith, 1993, 1994, 1996), and I am not alone

in this assessment (cf. Boorse, 2002; Millikan, 2002; Schwartz, 2002; Collier,

2000; see also Arp, 2006b). Ultimately, I want to maintain that these two

accounts need not be in competition and, actually, can complement one

another. Further, as will be shown later, my hypothesis concerning the

emergence of scenario visualization will rely upon parts of both of these

accounts.

Cummins explains the function of some trait in terms of the role it plays

in maintaining the overall organization and survival of the organism in

its present state. Traits only have functions in relation to other traits within

the organization of the system as a whole. The function of the heart is a standard case utilized in discussions. Thus, the heart’s function is to pump

blood because this fulfi lls its causal role in relation to the organization

of the animal as a whole. There are two things to notice about this

account.

First, a trait can have—or could have had, or could have—any variety of

functions in relation to the organizational whole. In other words, a single

part could have a variety of functions, depending upon the organization

of the system. The heart could have fi ltered urine if the overall organization

of the animal were different in the past, or it could be co-opted to

perform some other function if the overall organization of the animal

changed in the future. Thus, there is a certain fl exibility or malleability in

some trait’s functionality. I fi nd this to be a virtue of Cummins’ position,

as it is now common knowledge that certain traits may actually be exaptations

or preaptations, rather than adaptations. Although, as we will see,

the fl exibility in a trait’s functionality makes it diffi cult to defi ne exactly

what a trait’s present function in the overall organization of an organism

actually is.

Second, Cummins purposely develops his account independent of any

evolutionary factors that would contribute to the historical origin of a

trait’s functionality. According to Cummins, for a trait to perform some

function, the basic structural components of the trait must already be

present so that it can perform that specifi c function in relation to the

organization of the organism as a whole. In other words, because the function

of some trait cannot precede the presence of that trait, and only comes

to have a function in relation to the contribution it makes to the organization

of the organism as a whole in its present state, it makes no sense to

speak about an evolutionary history with respect to that trait. Thus, hearts

pump blood because of their contribution to the overall maintenance of

the animal’s life, not because they conferred a survival-enhancing capacity

in the animal’s evolutionary ancestry. Whereas the fl exibility and/or malleability

exhibited in a trait’s functionality acts as a virtue of Cummins’

position, I fi nd the neglect of a trait’s history to be a vice of Cummins’

position, as I will demonstrate in a moment.

Conversely, according to the Griffi ths/Godfrey-Smith modern history

account, the function of some trait or process X in an organism is defi ned

by what the trait was naturally selected for doing in the organism’s species’

recent past. This is to say that past advantages to the organism are what

defi ne functions. Griffi ths and Godfrey-Smith add the qualifi cation that

the past advantage must be a recent addition in the species’ history, stemming

from the fact that the original selection for any trait may have favored an entirely different effect than the one that counts as the trait’s

current function. Thus, for example, the heart pumps blood or the kidney

fi lters urine in a cat because these traits enabled the cat’s most recent

ancestors to survive. The hearts that pumped blood and the kidneys that

fi ltered urine most adequately were naturally selected for as a trait in the

feline species, and cats today have hearts and kidneys that function so as

to pump blood and fi lter urine precisely because of this fi tness. Ayala (1998,

p. 45) favors a historical account of functions and notes that some trait is

functional “if it contributes to the reproductive effi ciency of the organism

itself, and if such contribution accounts for the existence of the structure

or process.” Ayala (1998, p. 40) further notes that these traits in organisms

are “biological adaptations. They have arisen as a result of the process of

natural selection. The adaptations of organisms—whether organs, homoeostatic

mechanisms, or patterns of behavior—are explained [functionally]

in that their existence is accounted for in terms of their contribution to

the reproductive fi tness of the population.”

According to Cummins, the function of some trait X has to do with the

trait’s contribution to some greater capacity or propensity of the organism

to survive rather than its evolutionary history. From Cummins’ perspective,

to say that the heart’s function is simply its adaptive ability to pump

blood—as proponents of the Griffi ths/Godfrey-Smith modern history

account want to do—is to miss other relevant effects of the heart that

contribute to the overall survival value of the organism. The heart produces

sound, weighs a certain amount, and has a certain structure, and these are

all factors besides its pumping of blood that supposedly contribute to the

overall maintenance, survival, and propensity for survival in the organism

(cf. Bigelow & Pargetter, 1998; Bechtel, 1989; Allen, Bekoff, & Lauder, 1998;

Staddon, 1987; Horan, 1989).

Unfortunately, the problem with this description of functions is that it

is too broad. Cummins fails to distinguish effects it is the function of a

trait or organ to produce from those it is not its function to produce.

Almost any effect can be understood as contributing to the survival value

of an organism on this interpretation of function. In other words, describing

functions purely in terms of organizational capacities and propensities

for survival puts us in the awkward position of not being able to distinguish

between the salient and nonsalient purposive features of traits. Concerning

the example of the heart, Cummins wants to say that weighing a certain

amount, producing sound, having a certain structure, and pumping blood

are all potential functions of the heart. However, there does not seem to

be any signifi cant evidence or theoretical precedent to show that the heart’s producing of sounds or weighing a certain amount per se confers

any propensity for survival in an organism.

There is a further problem with Cummins’ account. Recall that a trait

can have—or could have had, or could have—any variety of functions in

relation to the organizational whole. A single part could have a variety of

functions, depending upon the organization of the system. However, it

does not seem right to speak about the function of some trait by reference

to a propensity that will confer future advantage, because the future has

not occurred for us to know if in fact the advantage will be conferred. As

Cummins himself acknowledges, it is possible that a trait adapted to function

in an environment now may be co-opted for a different use in that

same environment later. The ability to fl y in birds may have come about

in this way. Feathers probably were selected for in archaic birds to protect

them from the cold but eventually became co-opted for fl ight (Feduccia,

1996; Ostrom, 1979; Gould & Vrba, 1982).

Also consistent with Cummins’ account is the fact that the environment

may change, possibly making it such that a trait that functioned in the

previous environment will become nonfunctional in the present environment.

It is likely that the existence of vestigial organs testifi es to this kind

of change (Berra, 1990). Further, it may be possible that an environment

changes, and a trait adapted to function in the previous environment gets

altered and co-opted for a different use in the present environment. This

may be the way in which conscious decision making arose in humans

when the hominin line was forced to move from jungle to savanna in

Africa during the Pleistocene epoch (see Tattersall, 2001, 2002; Arp, 2005a,

2006a). Thus, we never really can say that a trait will have some future

functioning advantage. At best, we look at its present functioning in relation

to some environment and surmise about its past functioning—again,

recent past, as per the qualifi cation of Griffi ths and Godfrey-Smith—in

some similar or dissimilar environment.

So, Cummins seems to do at least two things that are misguided. First,

he inappropriately neglects the role that recent evolutionary history plays

in a trait’s functional development. Second, in the tongue-in-cheek words

of Achinstein (1977, p. 344), he “saddles us with a bevy of unwanted functions.”

Most biologists would agree with the Griffi ths/Godfrey-Smith

account and want to say that the proper function of the heart—given its

recent selection in evolutionary history—is to pump blood, and only in an

ancillary way is the heart’s function to produce sounds, weigh a certain

amount, or have a certain structure (see Neander, 1999; Ayala, 1972; Sober,

1993; Millikan, 1984, 2002; Mayr, 1993; Dawkins, 1986; Ruse, 1973).

Nonetheless, it seems to me that the accounts of Cummins and Griffi ths/

Godfrey-Smith can complement and be made compatible with one another.

The part of Cummins’ account of functions that seems correct has to do

with the function of a trait understood as both (1) contributing to the

overall organizational structure of the organic system and (2) being fl exible

or malleable enough to be co-opted for some other function. These qualities

can complement the idea that the functioning of a trait must arise

within the context of an evolutionary history. There is nothing incompatible

or contradictory in maintaining that some trait functions so as to

contribute to the general organization of some organism’s structure while

at the same time describing the emergence of that trait’s functional contribution

to the organizational structure of the organism by reference to a

recent evolutionary history. So too, there is nothing incompatible or contradictory

in maintaining that some trait is malleable enough to be utilized

for several potential functions while at the same time describing the present

status of that trait in reference to a recent evolutionary history. Given that

structure, organization, operational fl exibility, function, and evolutionary

history are all factors to be considered in an organism’s makeup, we should

expect that the traits of an organism function the way they do because

such traits presently contribute to the overall organization of the organism

(Cummins) as well as were selected for in the organism’s species’ recent

ancestry (Griffi ths/Godfrey-Smith).

In attacking historical accounts of function, including the modern

history account, Bigelow & Pargetter (1998) entertain the thought experiment

that the tenets of evolutionary biology may turn out to be false. For

example, your grandparents could have been put together randomly out

of swamp material, and hence, you would have no evolutionary history to

speak of. If that happened to be the case, then the backward-looking, historical

explanation of function would lose its very backward support! I will

risk sounding ad hominem and point out—as Churchland (1993, p. 746)

has done—that this kind of thought experiment suffers from the fl aw of

most armchair philosophical thought experiments, namely, “too much

thought and not enough experiment!” One of the marks of good scientifi

c theorizing has to do with being open to the possibility that a wellestablished

theory may one day be debunked by evidence that disconfi rms

the theory. In other words, good science entails the falsifi cation of theory.

The swamp grandparents thought experiment hints at this falsifi cation

with respect to evolutionary theory. However, as Millikan (1989, 2002)

and Neander (1991) affi rm, there is enough evidence to suggest that we

should not abandon evolutionary theory just yet, despite such a thought experiment. I will point to a lot of this evidence when I build my case for

the evolution of the brain, visual system, and scenario visualization later

in this book. The work of biologists, neuroscientists, geologists, archeologists,

psychologists, philosophers, anthropologists, and zoologists, among

legions of other thinkers and researchers, establishes evolutionary theory

on a fi rm footing.

Despite the divergent views concerning the defi nition of a biological

function, both proponents of Cummins-style functions and proponents of

Griffi ths/Godfrey-Smith-style functions agree that evolution has taken

place and continues to take place. Given that evolution is understood as

the result of the complex workings of such factors as differential reproduction,

artifi cial selection, sexual selection, mutation, genetic drift, and

genetic recombination in a historical context, this historicity lends further

support to approaching the functions of traits from the perspective of the

modern history account.

The modern history account gives elucidation to Mayr’s (1996, p. 103)

description of organisms as “operating on the basis of historically acquired

programs of information.” Crudely and simply put, the information that

organisms acquire is to be found in the particular genetic code received

from their parent(s). However, as Plotkin (1997, p. 1) has observed, “nothing

in biology makes complete sense [italics mine] except in light of evolution.”

Therefore, any description of biological phenomena must include an evolutionary

perspective to be considered as a fully explanatory description

of the biological phenomena under investigation.

The account of functions I endorse is an evolutionary adaptationist

explanation. It is well-known that the primary mechanism of evolution is

natural selection. Through natural selection, environmental infl uences

affect populations of organisms, and the chance that benefi cial traits will

dominate in successor generations is increased by the adaptive ability and

reproductive success of individuals possessing optimal genetic variants.

When a trait contributes to the fi tness of the organism in its current environment,

it is said to be functionally adaptive. For example, as was mentioned

in the previous chapter, the various kinds of fi nches that Darwin

described on the Galapagos Islands have different beak structures as a

functionally adaptive response to their particular food source. Dawkins

(1986, p. 178) is correct in maintaining that adaptations “affect every part

of the body, its shape and colour, its internal organs, its behaviour, and

the chemistry of its cells.”

A Cummins-style explanation is a necessary ingredient in the description

of a biological function, given the dual emphasis placed upon a trait’s role in the overall organization of the organism and the fl exibility associated

with a trait’s function. However, a complete account of the function of

some trait requires an explanation of how the trait came to be useful for

the organism. Thinkers who utilize the evolutionary adaptationist methodology

view traits as adaptations that have evolved due to past contributions

to the fi tness of the organism in some environment. Sober (1993,

p. 83) is representative of this kind of thinking, and he links adaptation

to a historical account in a clear fashion: “To say that a trait is an ‘adaptation’

is to comment not on its current utility but on its history. To say that

the mammalian heart is (now) an adaptation for pumping blood is to say

that mammals now have hearts because ancestrally, having a heart conferred

a fi tness advantage; the trait evolved because there was selection for

having a heart, and hearts were selected because they pump blood.”

The biologist in me takes it as a real given that evolution has occurred

and that accounts of the functioning of traits and processes require an

evolutionary explanation. The philosopher of biology in me is not sure if

there is anything real out there to speak of as having a function; however,

my as-if realism tells me that it makes sense to proceed as if there were

realities out there with functions based in an evolutionary history. Thus,

in the end, given my endorsement of the modern history account as

necessary to explain how it is that organisms have become hierarchically

organized functioning systems, I take Plotkin’s claim that a biological

phenomenon only makes complete sense in light of evolutionary theory

seriously.

In the fi rst chapter, I described an organism as a living entity, the components

of which are hierarchically organized in subsystems and processes

operating so as to achieve particularized and generalized homeostasis. The

subsystems and processes possess certain properties, including abilities to

fl exibly exchange data, convert data to information in a selection process,

integrate information, and process information from environments. In this

chapter, I endorsed a form of metaphysical emergence whereby entities,

properties, or substances arise out of more fundamental entities, properties,

or substances and yet are not wholly reducible to them. I argued for this

form of emergence based upon HOV, namely, the fact that the subsystems

and processes of organisms coordinate their functions so as to produce

particularized and generalized homeostasis. I also argued for an epistemological

form of emergence that is rooted in the language of teleology and

as-if realism.

In completing my account of organisms understood as hierarchically

organized systems, I argued that the components of organisms function the way they do as a result of recent past adaptive advantages. The evolutionary

explanation I endorsed is a type of causal explanation of how it is

that the recent history of a trait has had certain adaptive effects, which

have facilitated the selection of that trait. We will see in subsequent chapters

that the brain and visual system, as well as the psychological phenomena

emerging from brain processes that enable humans to solve problems

creatively, exhibit similar biological properties and abide by the same

evolutionary principles laid out and explained in these last two chapters.

In the last two sections of this chapter, I have argued for a nomological

form of metaphysical emergence, in terms of HOV, and I began to argue

for a representational form of epistemological emergence, in terms of an

as-if realism. In the fi nal section of this chapter, I will continue to argue

for both forms of emergence by an investigation of function (also see Arp,

2006b). That organisms have functions and that we must describe—or, at

least, as-if describe—the actual functioning of those organisms are central

to my metaphysical and epistemological versions of emergence. Further,

it is essential to my project that I explain and defend a description of functions

because my hypothesis concerning scenario visualization depends

upon certain functional mechanisms of the mind having evolved to solve

specifi c problems encountered in various Pleistocene environments. As a

biologist (of sorts), the realist in me says that organisms really do have

functions. As a philosopher of biology, the antirealist in me realizes that no

one can know, for sure, if organisms really do have functions, principally

because of the veil of perceptions that mediates between the mind and

world (if there is a world). However, we still must proceed to describe

organisms as if they have functions because (1) such a description is useful

and (2) it would be miraculous if our descriptions did not match up with

what we take to be the real functions of organisms as we perceive them.

Organisms, and the subsystems and processes that comprise them, tend

to operate in certain ways on a regular basis. This operational regularity

not only aids biologists in identifying certain traits but enables biology to

be considered an autonomous science with its own domain of laws (also

see the sections on function in Arp & Ayala, 2008; Arp & Rosenberg, 2008;

Terzis & Arp, 2008). Genetic information is communicated from parent

organism to offspring, mitochondria convert sugar to ATP, the heart pumps

blood, the medulla controls breathing, the complex interactions of the

subsystems and processes of multicellular organisms tend toward the maintenance

of a dynamic equilibrium (homeostasis)—all of this happens in

fairly predictable and reliable ways. Another way of describing the tendency

for processes and subsystems of organisms to operate in predictable

and reliable ways is in terms of the organism’s functions.

In the biological realm, a complete explanation of a trait seems to

include an explanation in terms of function. It is natural to ask of some

trait, What is its function? or What purpose in the organism does the particular

trait serve? or What is the goal of its activity? Thus, for example,

to explain the existence of the human heart as merely a red mass of tissue located in the chest cavity of the torso that pulsates and reverberates

according to the natural laws of physics and chemistry strikes us as an

incomplete explanation. However, as has been argued quite convincingly

by Mayr (1996), Walters (1998), Lennox (1993), Sober (1993), Gould (1977),

Sterelny (2001), Dawkins (1986), and others, biology has its own legitimate

scientifi c principles and terminology (cf. Machamer, Darden, & Craver,

2000; Tabery, 2004; Glennan, 1996). In line with a biological explanation,

it seems we can talk about the heart’s function or purpose in the human

body as an organ that pumps blood. An important part of explaining the

heart is to say that it exists so as to pump blood, or that pumping blood

is the heart’s operation or function.

Beginning in the 1960s with Nagel (1961) and Hempel (1965), philosophers

of science and biology have not only proposed a variety of defi nitions

for the term function but have attempted to specify the conditions

where it may be appropriate to predicate functions of natural and artifi cial

things, and to appeal to these functions in explanations (see Cummins,

1975, 2002; Wright, 1973, 1976; Mayr, 1982, 1988; Millikan, 1984, 1989,

2002; Buller, 1999; Ayala, 1998; Mahner & Bunge, 2001; Neander, 1991,

1999; Hardcastle, 1999; Perlman, 2004). Some thinkers, like Dennett (1987,

1995) and the Churchlands (1986, 1989), offer a promissory note that talk

of functions in biology will be eliminated altogether once a more accurate,

value-free system of terms is invented to describe the operational traits of

organisms. Searle (1990b, p. 414) seems to hold the similar view that functions

are solely the product of intentional and intensional minds, and that

functions are “never intrinsic but are always observer relative.” And Ruse

has affi rmed (1971, 1973) and reaffi rmed (2003) a position similar to that

of Dennett, the Churchlands, and Searle, namely, there is nothing like an

intrinsic or extrinsic ontological purpose or design to be found in nature.

Talk of functions is just that—talk of functions.

Yet Ruse and other thinkers admit something about the biological sciences

that Kant (1987) had admitted in his third Critique; with respect to

organisms, it is useful to think as if these entities have traits and processes

that function in goal-directed ways. Thinkers cannot seem to get around

Trivers’ (1985, p. 5) claim that “even the humblest creature, say, a virus,

appears organized to do something; it acts as if [italics mine] it is trying to

achieve some purpose” or Arnhart’s (1998, p. 245) observation that

“although the evolutionary process does not serve goals, the organisms

emerging from that process do. Darwin’s biology does not deny—rather,

it reaffi rms—the immanent teleology displayed in the striving of each

living being to fulfi ll its specifi c ends. . . . Reproduction, growth, feeding, healing, courtship, parental care for the young—these and many other

activities of organisms are goal-directed” (also see Bogdan, 1994; Stout,

1996; FitzPatrick, 2000; Sterelny & Griffi ths, 1999; Arp, 1998, 1999, 2002,

2006b).

In the last section, I argued for an as-if realism. This approach acknowledges

that we could never know the nature of anything outside of our

own perceptions for certain. At the same time, as-if realism affi rms that

we must act as if there is a reality out there to be understood and

described; it would be miraculous if there were not a real world out

there to which our perceptions and ideas correspond, given the fecundity

of our scientifi c, philosophic, and other research endeavors. With this

as-if realism in mind, we can still attempt to describe the functions of

the traits and processes of organisms. Thus, I am in agreement with Ruse

and other thinkers who do not attempt to explain away descriptions of

the functions of traits. The as-if realism I endorse works together nicely

with this as-if description of the functioning of biological traits—the

as-if realist in me acknowledges that organisms act as if they are organized

so as to function in specifi c ways. In fact, as noted already, it is

essential to my project that I lay out and defend a description of functions

because, as will be shown in later chapters, my hypothesis concerning

scenario visualization depends upon certain functional mechanisms

of the mind having evolved to solve specifi c problems encountered in

various Pleistocene environments.

There are several views concerning the appropriate defi nition of function

for biological matters, and Perlman (2004) has done a fi ne job of laying

out all of these views as envisioned in the history of Western philosophy.

The two views of function with respect to living things that, in my estimation,

have the most credibility are Cummins’ (1975, 2002) organizational

account and the Griffi ths/Godfrey-Smith modern history account (Griffi ths,

1992, 1993, 1996; Godfrey-Smith, 1993, 1994, 1996), and I am not alone

in this assessment (cf. Boorse, 2002; Millikan, 2002; Schwartz, 2002; Collier,

2000; see also Arp, 2006b). Ultimately, I want to maintain that these two

accounts need not be in competition and, actually, can complement one

another. Further, as will be shown later, my hypothesis concerning the

emergence of scenario visualization will rely upon parts of both of these

accounts.

Cummins explains the function of some trait in terms of the role it plays

in maintaining the overall organization and survival of the organism in

its present state. Traits only have functions in relation to other traits within

the organization of the system as a whole. The function of the heart is a standard case utilized in discussions. Thus, the heart’s function is to pump

blood because this fulfi lls its causal role in relation to the organization

of the animal as a whole. There are two things to notice about this

account.

First, a trait can have—or could have had, or could have—any variety of

functions in relation to the organizational whole. In other words, a single

part could have a variety of functions, depending upon the organization

of the system. The heart could have fi ltered urine if the overall organization

of the animal were different in the past, or it could be co-opted to

perform some other function if the overall organization of the animal

changed in the future. Thus, there is a certain fl exibility or malleability in

some trait’s functionality. I fi nd this to be a virtue of Cummins’ position,

as it is now common knowledge that certain traits may actually be exaptations

or preaptations, rather than adaptations. Although, as we will see,

the fl exibility in a trait’s functionality makes it diffi cult to defi ne exactly

what a trait’s present function in the overall organization of an organism

actually is.

Second, Cummins purposely develops his account independent of any

evolutionary factors that would contribute to the historical origin of a

trait’s functionality. According to Cummins, for a trait to perform some

function, the basic structural components of the trait must already be

present so that it can perform that specifi c function in relation to the

organization of the organism as a whole. In other words, because the function

of some trait cannot precede the presence of that trait, and only comes

to have a function in relation to the contribution it makes to the organization

of the organism as a whole in its present state, it makes no sense to

speak about an evolutionary history with respect to that trait. Thus, hearts

pump blood because of their contribution to the overall maintenance of

the animal’s life, not because they conferred a survival-enhancing capacity

in the animal’s evolutionary ancestry. Whereas the fl exibility and/or malleability

exhibited in a trait’s functionality acts as a virtue of Cummins’

position, I fi nd the neglect of a trait’s history to be a vice of Cummins’

position, as I will demonstrate in a moment.

Conversely, according to the Griffi ths/Godfrey-Smith modern history

account, the function of some trait or process X in an organism is defi ned

by what the trait was naturally selected for doing in the organism’s species’

recent past. This is to say that past advantages to the organism are what

defi ne functions. Griffi ths and Godfrey-Smith add the qualifi cation that

the past advantage must be a recent addition in the species’ history, stemming

from the fact that the original selection for any trait may have favored an entirely different effect than the one that counts as the trait’s

current function. Thus, for example, the heart pumps blood or the kidney

fi lters urine in a cat because these traits enabled the cat’s most recent

ancestors to survive. The hearts that pumped blood and the kidneys that

fi ltered urine most adequately were naturally selected for as a trait in the

feline species, and cats today have hearts and kidneys that function so as

to pump blood and fi lter urine precisely because of this fi tness. Ayala (1998,

p. 45) favors a historical account of functions and notes that some trait is

functional “if it contributes to the reproductive effi ciency of the organism

itself, and if such contribution accounts for the existence of the structure

or process.” Ayala (1998, p. 40) further notes that these traits in organisms

are “biological adaptations. They have arisen as a result of the process of

natural selection. The adaptations of organisms—whether organs, homoeostatic

mechanisms, or patterns of behavior—are explained [functionally]

in that their existence is accounted for in terms of their contribution to

the reproductive fi tness of the population.”

According to Cummins, the function of some trait X has to do with the

trait’s contribution to some greater capacity or propensity of the organism

to survive rather than its evolutionary history. From Cummins’ perspective,

to say that the heart’s function is simply its adaptive ability to pump

blood—as proponents of the Griffi ths/Godfrey-Smith modern history

account want to do—is to miss other relevant effects of the heart that

contribute to the overall survival value of the organism. The heart produces

sound, weighs a certain amount, and has a certain structure, and these are

all factors besides its pumping of blood that supposedly contribute to the

overall maintenance, survival, and propensity for survival in the organism

(cf. Bigelow & Pargetter, 1998; Bechtel, 1989; Allen, Bekoff, & Lauder, 1998;

Staddon, 1987; Horan, 1989).

Unfortunately, the problem with this description of functions is that it

is too broad. Cummins fails to distinguish effects it is the function of a

trait or organ to produce from those it is not its function to produce.

Almost any effect can be understood as contributing to the survival value

of an organism on this interpretation of function. In other words, describing

functions purely in terms of organizational capacities and propensities

for survival puts us in the awkward position of not being able to distinguish

between the salient and nonsalient purposive features of traits. Concerning

the example of the heart, Cummins wants to say that weighing a certain

amount, producing sound, having a certain structure, and pumping blood

are all potential functions of the heart. However, there does not seem to

be any signifi cant evidence or theoretical precedent to show that the heart’s producing of sounds or weighing a certain amount per se confers

any propensity for survival in an organism.

There is a further problem with Cummins’ account. Recall that a trait

can have—or could have had, or could have—any variety of functions in

relation to the organizational whole. A single part could have a variety of

functions, depending upon the organization of the system. However, it

does not seem right to speak about the function of some trait by reference

to a propensity that will confer future advantage, because the future has

not occurred for us to know if in fact the advantage will be conferred. As

Cummins himself acknowledges, it is possible that a trait adapted to function

in an environment now may be co-opted for a different use in that

same environment later. The ability to fl y in birds may have come about

in this way. Feathers probably were selected for in archaic birds to protect

them from the cold but eventually became co-opted for fl ight (Feduccia,

1996; Ostrom, 1979; Gould & Vrba, 1982).

Also consistent with Cummins’ account is the fact that the environment

may change, possibly making it such that a trait that functioned in the

previous environment will become nonfunctional in the present environment.

It is likely that the existence of vestigial organs testifi es to this kind

of change (Berra, 1990). Further, it may be possible that an environment

changes, and a trait adapted to function in the previous environment gets

altered and co-opted for a different use in the present environment. This

may be the way in which conscious decision making arose in humans

when the hominin line was forced to move from jungle to savanna in

Africa during the Pleistocene epoch (see Tattersall, 2001, 2002; Arp, 2005a,

2006a). Thus, we never really can say that a trait will have some future

functioning advantage. At best, we look at its present functioning in relation

to some environment and surmise about its past functioning—again,

recent past, as per the qualifi cation of Griffi ths and Godfrey-Smith—in

some similar or dissimilar environment.

So, Cummins seems to do at least two things that are misguided. First,

he inappropriately neglects the role that recent evolutionary history plays

in a trait’s functional development. Second, in the tongue-in-cheek words

of Achinstein (1977, p. 344), he “saddles us with a bevy of unwanted functions.”

Most biologists would agree with the Griffi ths/Godfrey-Smith

account and want to say that the proper function of the heart—given its

recent selection in evolutionary history—is to pump blood, and only in an

ancillary way is the heart’s function to produce sounds, weigh a certain

amount, or have a certain structure (see Neander, 1999; Ayala, 1972; Sober,

1993; Millikan, 1984, 2002; Mayr, 1993; Dawkins, 1986; Ruse, 1973).

Nonetheless, it seems to me that the accounts of Cummins and Griffi ths/

Godfrey-Smith can complement and be made compatible with one another.

The part of Cummins’ account of functions that seems correct has to do

with the function of a trait understood as both (1) contributing to the

overall organizational structure of the organic system and (2) being fl exible

or malleable enough to be co-opted for some other function. These qualities

can complement the idea that the functioning of a trait must arise

within the context of an evolutionary history. There is nothing incompatible

or contradictory in maintaining that some trait functions so as to

contribute to the general organization of some organism’s structure while

at the same time describing the emergence of that trait’s functional contribution

to the organizational structure of the organism by reference to a

recent evolutionary history. So too, there is nothing incompatible or contradictory

in maintaining that some trait is malleable enough to be utilized

for several potential functions while at the same time describing the present

status of that trait in reference to a recent evolutionary history. Given that

structure, organization, operational fl exibility, function, and evolutionary

history are all factors to be considered in an organism’s makeup, we should

expect that the traits of an organism function the way they do because

such traits presently contribute to the overall organization of the organism

(Cummins) as well as were selected for in the organism’s species’ recent

ancestry (Griffi ths/Godfrey-Smith).

In attacking historical accounts of function, including the modern

history account, Bigelow & Pargetter (1998) entertain the thought experiment

that the tenets of evolutionary biology may turn out to be false. For

example, your grandparents could have been put together randomly out

of swamp material, and hence, you would have no evolutionary history to

speak of. If that happened to be the case, then the backward-looking, historical

explanation of function would lose its very backward support! I will

risk sounding ad hominem and point out—as Churchland (1993, p. 746)

has done—that this kind of thought experiment suffers from the fl aw of

most armchair philosophical thought experiments, namely, “too much

thought and not enough experiment!” One of the marks of good scientifi

c theorizing has to do with being open to the possibility that a wellestablished

theory may one day be debunked by evidence that disconfi rms

the theory. In other words, good science entails the falsifi cation of theory.

The swamp grandparents thought experiment hints at this falsifi cation

with respect to evolutionary theory. However, as Millikan (1989, 2002)

and Neander (1991) affi rm, there is enough evidence to suggest that we

should not abandon evolutionary theory just yet, despite such a thought experiment. I will point to a lot of this evidence when I build my case for

the evolution of the brain, visual system, and scenario visualization later

in this book. The work of biologists, neuroscientists, geologists, archeologists,

psychologists, philosophers, anthropologists, and zoologists, among

legions of other thinkers and researchers, establishes evolutionary theory

on a fi rm footing.

Despite the divergent views concerning the defi nition of a biological

function, both proponents of Cummins-style functions and proponents of

Griffi ths/Godfrey-Smith-style functions agree that evolution has taken

place and continues to take place. Given that evolution is understood as

the result of the complex workings of such factors as differential reproduction,

artifi cial selection, sexual selection, mutation, genetic drift, and

genetic recombination in a historical context, this historicity lends further

support to approaching the functions of traits from the perspective of the

modern history account.

The modern history account gives elucidation to Mayr’s (1996, p. 103)

description of organisms as “operating on the basis of historically acquired

programs of information.” Crudely and simply put, the information that

organisms acquire is to be found in the particular genetic code received

from their parent(s). However, as Plotkin (1997, p. 1) has observed, “nothing

in biology makes complete sense [italics mine] except in light of evolution.”

Therefore, any description of biological phenomena must include an evolutionary

perspective to be considered as a fully explanatory description

of the biological phenomena under investigation.

The account of functions I endorse is an evolutionary adaptationist

explanation. It is well-known that the primary mechanism of evolution is

natural selection. Through natural selection, environmental infl uences

affect populations of organisms, and the chance that benefi cial traits will

dominate in successor generations is increased by the adaptive ability and

reproductive success of individuals possessing optimal genetic variants.

When a trait contributes to the fi tness of the organism in its current environment,

it is said to be functionally adaptive. For example, as was mentioned

in the previous chapter, the various kinds of fi nches that Darwin

described on the Galapagos Islands have different beak structures as a

functionally adaptive response to their particular food source. Dawkins

(1986, p. 178) is correct in maintaining that adaptations “affect every part

of the body, its shape and colour, its internal organs, its behaviour, and

the chemistry of its cells.”

A Cummins-style explanation is a necessary ingredient in the description

of a biological function, given the dual emphasis placed upon a trait’s role in the overall organization of the organism and the fl exibility associated

with a trait’s function. However, a complete account of the function of

some trait requires an explanation of how the trait came to be useful for

the organism. Thinkers who utilize the evolutionary adaptationist methodology

view traits as adaptations that have evolved due to past contributions

to the fi tness of the organism in some environment. Sober (1993,

p. 83) is representative of this kind of thinking, and he links adaptation

to a historical account in a clear fashion: “To say that a trait is an ‘adaptation’

is to comment not on its current utility but on its history. To say that

the mammalian heart is (now) an adaptation for pumping blood is to say

that mammals now have hearts because ancestrally, having a heart conferred

a fi tness advantage; the trait evolved because there was selection for

having a heart, and hearts were selected because they pump blood.”

The biologist in me takes it as a real given that evolution has occurred

and that accounts of the functioning of traits and processes require an

evolutionary explanation. The philosopher of biology in me is not sure if

there is anything real out there to speak of as having a function; however,

my as-if realism tells me that it makes sense to proceed as if there were

realities out there with functions based in an evolutionary history. Thus,

in the end, given my endorsement of the modern history account as

necessary to explain how it is that organisms have become hierarchically

organized functioning systems, I take Plotkin’s claim that a biological

phenomenon only makes complete sense in light of evolutionary theory

seriously.

In the fi rst chapter, I described an organism as a living entity, the components

of which are hierarchically organized in subsystems and processes

operating so as to achieve particularized and generalized homeostasis. The

subsystems and processes possess certain properties, including abilities to

fl exibly exchange data, convert data to information in a selection process,

integrate information, and process information from environments. In this

chapter, I endorsed a form of metaphysical emergence whereby entities,

properties, or substances arise out of more fundamental entities, properties,

or substances and yet are not wholly reducible to them. I argued for this

form of emergence based upon HOV, namely, the fact that the subsystems

and processes of organisms coordinate their functions so as to produce

particularized and generalized homeostasis. I also argued for an epistemological

form of emergence that is rooted in the language of teleology and

as-if realism.

In completing my account of organisms understood as hierarchically

organized systems, I argued that the components of organisms function the way they do as a result of recent past adaptive advantages. The evolutionary

explanation I endorsed is a type of causal explanation of how it is

that the recent history of a trait has had certain adaptive effects, which

have facilitated the selection of that trait. We will see in subsequent chapters

that the brain and visual system, as well as the psychological phenomena

emerging from brain processes that enable humans to solve problems

creatively, exhibit similar biological properties and abide by the same

evolutionary principles laid out and explained in these last two chapters.