the biology of the future and the future of biology

The Biology of the Future and the Future of Biology

Steven P. R. Rose1,2

1Brain and Behaviour Research Group, Department of Biology, The Open University, Milton Keynes MK7 6AA, UK

2Department of Anatomy and Developmental Biology, University College London, London, UK

Constraints on knowing nature

We view and interpret the world around us, bothnatural and cultural, through perceptual and cog-nitive spectacles of our own construction. Thenatural sciences claim that their methods, ofhypothesis, observation and experiment, permitsomething approximating to a true representationof the material reality that surrounds us to beachieved. However, for several decades now,philosophers, historians and sociologists of sciencehave been pointing to the ways in which ourscientific knowledge is socially, culturally andhistorically constructed—that is, it offers at best aconstrained interpretation of the material world. Afirst such constraint is provided by the very con-struction of our brain and the biology of our per-ceptual processes. To this I would add that brainsdo not exist in isolation from bodies: how we per-ceive the world is affected by our hormonal,immunological and general physiological state.And we perceive the world in the way that we dobecause our visual system is capable of sensingonly a limited range of wavelengths, our massand volume give us a particular relationship togravitational forces not shared for instance bybacteria or beetles, or by whales or elephants. Oursense of the temporality of events is shaped by thefact that we may live for anything up to and noweven beyond a century. Bacteria divide every20 minutes or so, mayflies live for a day, redwoodtrees for thousands of years. Human technologiescan and do enable us to escape these structuraland temporal limitations, to observe in the infraredor ultraviolet, weigh atoms and measure time inanything from nanoseconds to light years. Yeteven when considering the inconceivably small ordistant, we do so by scales that relate to ourhuman condition: the measure of man is man.

But there are other constraints that transcend ourmere biology. One—which is where the sciencesdiffer most from the arts and humanities—is that

we are not free to offer interpretations that ourobservations of, and experiments on, the externalworld disconfirm. A second is defined by the limitsof our available technologies.† Until the means ofcircumnavigating the earth were available, it wasa legitimate approximation to the truth to maintainthat the earth was flat. Until Lavoisier weighed theproducts of combustion, phlogiston theory was asgood as oxygen theory. Until microscopes revealedthe internal constituents of living tissue it waslegitimate to regard cells as composed of homo-geneous protoplasm.

But a third and equally important constraint isthat resulting from the very social and historicalnature of the scientific enterprise itself. The waysin which we view the world, the types of experi-ments we conceive and evidence we accept, thetheories we construct, are far from being culturallyfree. This means that we cannot understand thecurrent shape of biological thinking withoutreference to the history of our own discipline. Thisshould not surprise us. After all in a famousaphorism, known to all biologists, the greatevolutionist Theodosius Dobzhansky pointed outthat nothing in biology made sense except in thecontext of evolution. I would want merely tobroaden that statement, in ways that will becomeapparent in the course of this paper, to read“nothing in biology makes sense except in thecontext of history”—by which I mean evolutionaryhistory, developmental history, and the history ofour own subject.1 This does not imply a simpleprogressivism; new knowledge claims may wellbe but are not necessarily, “better” representationsof the material world than prior ones.

0022-2836/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved

This is an edited version of a paper first appeared inPerspectives in Biology and Medicine, 44(4), 473–484;reprinted by permission of the Johns Hopkins UniversityPress.

† Note that, unlike many commentators, I do notregard technology as fundamentally distinct fromscience. Whether knowledge and practice are scientific ortechnological depends on the use for which they areemployed, not something intrinsically different. Fordiscussion of this argument see e.g.: Rose, H. & Rose,S. P. R. (1969). Science and Society, (Penguin,Harmondsworth); Rose, S. P. R. (1993). The Making ofMemory, (Transworld/Bantam, Uxbridge).

doi:10.1016/S0022-2836(02)00332-7 available online at http://www.idealibrary.com onBw

J. Mol. Biol. (2002) 319, 877–884

The power of reductionist thinking and theplurality of biological explanation

For reasons that it would take me too far outsidemy theme to explore here, throughout its post-Cartesian and Newtonian history western sciencehas seen physics as its explanatory model. Themore pluralistic, pre-scientific world gave way toone in which all our day-to-day experimental rich-ness of colour, and sound, of love and anger, cameto be seen as secondary qualities underlyingwhich there were the changeless particles, wavesand forces of the physicists’ world. The task ofother sciences, chemistry, biology and laterpsychology, sociology and economics, was tomake themselves as like physics as possible. Theirqualities needed to be reduced to the true, quanti-tative “hard science” of physics. As philosopherThomas Nagel claims,2 other sciences describethings, reductionism explains them. The molecularbiologist James Watson, following Rutherford, putit more bluntly: ”there is only one science, physics;everything else is social work”.†

A further problem for biology, perhaps becauseits subject matter is so complex, is that there hasbeen a continuing tendency to understand livingprocesses and systems by metaphorising them tothe most advanced forms of current humanartefact. Many origin myths, including the Judaeo-Christian one, refer to humankind having beencreated from the dust or clay of the earth, like apotter’s wheel. The Greek Hero was said to havecreated lifelike robots using steam and waterpower. Hydraulic imagery persisted through theRenaissance (e.g. hearts as pumps; nerves as tubesfor conveying vital forces) to give way to electricaland magnetic ones in the 18th and 19th century.Brains became telegraph systems, then telephonesand now computers. Such metaphors are powerful,and may be helpful. But too often their seductivepowers blinker our capacity to see the world. As Iwill argue, brains are not computers, and genesare not selfish.

Let me offer a fable to demonstrate the limits ofreductionism in biology. Five biologists are on apicnic, when they see a frog jump into a pond.They fall to discussing why the frog has jumped.The first, a physiologist, describes the frog’s legmuscles and nervous system. The frog jumpsbecause impulses have travelled from the frog’sretina to its brain and thence down motor nervesto the muscle. The second, a biochemist, pointsout that the muscles are composed of actin andmyosin proteins—the frog jumps because of theproperties of these fibrous proteins that enablethem, driven by the energy of ATP, to slide pastone another. The third, a developmental biologist,describes the ontogenetic processes whereby thefertilised ovum divides, in due course forming thenervous system and musculature. The fourth, a

student of animal behaviour, points to a snake in atree above where the frog was sitting: the frogjumps to escape the snake. The fifth, an evolution-ist, explains the processes of natural selection thatensured that only those frog ancestors able both todetect snakes and jump fast enough to escapethem had a chance to survive and breed.

Five biologists, five very different types ofexplanation. Which is the right one? Answer: allof them are right, just different. The biochemist’sexplanation is the reductionist one, but it in noway eliminates the need for the others. Nor canwe envisage a research programme in which theother types of explanation would in due course allbe subsumed by either the biochemical, or theevolutionary one—despite the often quoted claimin the opening sentences of Wilson’s book onsociobiology,3 reiterated most recently in hisdemand for “consilience”.4 The most that we caninsist is that explanations in the different dis-courses should not contradict one another. As amaterialist, as all biologists must be, I am com-mitted to the view that we live in a world that isan ontological unity, but I must also accept anepistemological pluralism. As the philosopherMary Midgley puts it, neither the value of money,nor the rules of football, are collapsible intophysics; there is one world, but it is a big one.5

The shifting boundaries between natureand culture

Which type of explanation we prefer depends onthe purposes for which it is required. If we are con-cerned with diagnosing and treating diseases likemuscular dystrophy or myasthenia gravis, geneticand biochemical approaches may point the way.For others, like understanding why a personchooses to take a swim or a baby learns to walk,they are almost entirely useless. Far from such“lower level” accounts explaining, as Nagel wouldhave it, they at best merely describe, whilst thehigher and strictly irreducible accounts have themost explanatory power. Yet over the hundredand forty years since the publication of Darwin’sOrigin, biology’s pluralism has steadily beenrestricted. Physiology has been collapsed into bio-chemistry and biochemistry into chemistry andphysics. Skinner’s behaviouristic attempt to reducepsychology to physics and skip the interveningbiological level may have been rejected, but a newschool of “neurophilosophers” seeks to dismisswhat they call “folk psychology” in favour ofneurocomputation, in which brains are indeedreplaced by computers.6,7 Evolutionary biologyhas replaced study of the living world by abstractmathematical calculations about changes in thepopulation frequencies of individual selfishgenes.8 Indeed, the very phenotype, the livingorganism itself, has been emptied of any functionother than that of being the “lumbering robot”serving for the replication of its genes, to quoteDawkins. Molecular geneticists now see organisms

† In a debate with me at the Cheltenham book festivalin 1996.

878 Biology of the Future and the Future of Biology

as mere tools with which to probe gene function.They offer to predict our entire lifeline, our trajec-tory from birth to death, from the diseases we willdie of to the political parties we will vote for, thedrugs we will enjoy, our degree of job satisfactionand tendency to midlife divorce (For referencessee, Rose1). It would seem that reductionism hastriumphed.

What is clear is that developments in thesciences in general, and biology perhaps in particu-lar, have profoundly changed our concept both ofnature and of the boundaries between nature andculture. Take the concept of motherhood. Thatthere are both biological and social mothers haslong been recognised. But as feminists (e.g. theRose9) have emphasised, the new reproductivetechnologies have changed our understanding ofwhat it is to be “a mother”, so that there are nowgenetic mothers, surrogate (carry) mothers, socialmothers,… If human cloning becomes a reality,the concept of a mother will change yet again.

This reconceptualisation of what is ‘nature’ ismoving forward with great speed on the heels ofnew developments in genetics. Thus the neuro-sciences in collaboration with genetics are offeringto transform our understanding of human nature,turning what were once regarded as the resultseither of individual free will (or humanity’s sinfulnature—see for instance Milton’s Paradise Lost foran earlier orthodox theological account) into bio-medical matters. Adultery and cheating were oncesins; now evolutionary psychology tells us thatthese activities are the consequences of adaptationsduring humanity’s palaeolithic past (see Barkowet al.10 for an exposition of this argument and Rose& Rose11 for a critique). Alcoholism and violencewere social problems; now they are supposed tobe caused by abnormal genes. As an example, thelate John Kennedy explained that he didn’t“become” an alcoholic, it was something he was“born with”. Same sex love was once a sin, thena social problem; today there are claims that itis genetically caused (see Rose1 and referencestherein). Personal responsibility for our actions, agiven for both religious and humanist thinkers, isdissolved into double helix of DNA. Culturebecomes nature, whilst simultaneously biomedi-cine holds out the promise of transforming natureby genetic manipulation. Biology is Janus-faced. Itis determinist, as much predestinationist as anyreligious sect. But it is simultaneously Promethean,offering technology to conquer destiny. Thus thereis no unchanging nature out there constituted of“everything that is not man made artefact”. Theboundaries are not stable.

Yet there are limits to reductionism’s onwardmarch. Try as we may to collapse our sciences intogrand physical theories of everything (the physi-cists’ TOES and GUTS), such attempts must fail.The dream of writing a single equation that willembrace the world resembles more the search foran alchemical philosopher’s stone that will trans-mute base metal into gold and grant perpetual

youthful life to its possessor, or the cabbalisticbelief that there can be found a single mysticalsentence which will give its utterer almost god-like power over people and things. The ViennaSchool’s philosophical attempt in the 1930s toimpose a unity on the sciences—a unity builtupon physics—was similarly doomed to failure,and Wilson’s latest evocation of it is unlikely tosucceed.

Let me draw briefly on three areas of currentbiology to demonstrate the limits to reductionism’sdream and use these to point in the directions inwhich I believe that biology must move in its con-cept of nature in the new millennium: neuro-science, developmental and evolutionary biology.

Brains, minds and meaning

First, the brain. In order to understand how thehuman brain functions it is not sufficient to simplyextrapolate upward from the firing properties of itshundred billion neurons or the 1014 or so synapticconnections between them. Even a wiring diagramof their anatomical connections is insufficient. Thetemporal relationships between activity in distantbrain regions, coherent oscillatory processes, bind-ing mechanisms and doubtless as yet undiscoveredinteractions all make it necessary to consider thebrain as a system, not a mere assemblage of parts,perhaps to be understood using chaotic dynamictheory.12 I want to add one further point, and thatis to challenge the popular view that brains arecomputational, information processing devices, aclaim most forcefully made recently by Pinker13.

In Pinker’s view, following other evolutionarypsychologists, the brain/mind is not a general-purpose computer; rather it is composed of anumber of specific modules (for instance, a speechmodule, a number sense module, a face-recognition module, a cheat-detector module, andso forth). These modules have, it is argued,evolved quasi-independently during the evolutionof early humanity, and have persisted unmodifiedthroughout historical time, underlying theproximal mechanisms that traditional psychologydescribes in terms of motivation, drive, attentionand so forth. Whether such modules are morethan theoretical entities is unclear, at least to neuro-scientists. Indeed EP theorists such as Pinker go tosome lengths to make it clear that the “mentalmodules” they invent do not, or at least do notnecessarily, map onto specific brain structures. Butas Ellman et al.14 show, even if mental modules doexist they can as well be acquired as innate.

Modules or no, it is not adequate to reduce themind/brain to nothing more than a cognitive,“architectural” information processing machine.Brains/minds do not just deal with information.They are concerned with living meaning. In Howthe Mind Works, Pinker offers the example of afootprint as conveying information. My responseis to imagine Robinson Crusoe on his island, find-ing a footprint in the sand. First he has to interpret

Biology of the Future and the Future of Biology 879

the mark in the sand as that of a foot, and recognisethat it is not his own. But what does it mean tohim? Pleasure at the prospect of at last anotherhuman being to talk and interact with? Fear thatthis human may be dangerous? Memories of thesocial life of which he has been deprived for manyyears? A turmoil of thoughts and emotions withinwhich the visual information conveyed by the foot-print is embedded. The key here is emotion, for thekey feature which distinguishes brains/mindsfrom computers is their/our capacity to experienceemotion. Indeed, emotion is primary; affect asmuch as cognition is inextricably engaged in allbrain and mind processes, creating meaning out ofinformation—just one reason why brains aren’tcomputers. What is particularly egregious in thiscontext is Pinker’s oft-repeated phrase, “the archi-tecture of the mind”. Architecture, which impliesstatic structure, built to blueprints and thereafterstable, could not be a more inappropriate way toview the fluid dynamic processes whereby ourminds/brain develop and create order out of theblooming buzzing confusion of the world whichconfronts us moment by moment.

DNA and the cellular orchestra

Within the reductionist paradigm, developmentis the reading out of genetically encoded instruc-tions, present within the “master molecule”, DNA.Hence the popular references to DNA as being theblueprint of life, the code of codes, and so forth.The organism, the phenotype, is simply the vesselconstructed by the DNA in order to ensure its safereplication. This model misspeaks both therelationship of DNA to cellular processes ingeneral and the nature of development.

DNA itself is rather an inert molecule (hence thepossibility of recovering it intact from amber manytens of thousands of years old—and the plot ofJurassic Park). What brings it to life is the cell inwhich it is embedded. DNA cannot simply andunaided make copies of itself; it cannot therefore“replicate” in the sense that this term is usuallyunderstood. Replication—using one strand of thedouble helix of DNA to provide the template onwhich another can be constructed—requires anappropriately protected environment, the presenceof a wide variety of complex molecular precursors,a set of protein enzymes, and a supply of chemicalenergy. And even when DNA has been copiedfaithfully, the “read-out” into RNA and thenceinto protein, once though to be linear, is far frombeing so. Individual coding sequences of DNA aredistributed along chromosomes, punctuated bylong sequences (introns) of no known codingfunction. In humans some 98% of the DNA in thegenome comes into this non-coding category.Coding sequences are “read” by cellular mechan-isms, snipped out from the rest, spliced together,transcribed into RNA, edited.

The proteins they code for are then further pro-cessed and tailored by temporal and state regu-

lated cellular mechanisms quite distal to DNAitself. All these are provided in the complex meta-bolic web within which the myriad biochemicaland biophysical interactions occurring in each indi-vidual cell are stabilised.15 The famous “centraldogma” enunciated by Crick and on which gener-ations of biologists have been brought up, thatthere is a one way flow of information by which“DNA makes RNA makes protein” and that “onceinformation gets into the protein it can’t flow backagain”, was a superb simplification in the earlydays of molecular biology. But it simply isn’t trueany more. What is certain is that there are nomaster molecules in cellular processes. Even themetaphor of the cellular orchestra, which I haveused previously, is not adequate, as orchestrasrequire conductors. Better to see cells asmarvellously complex versions of string quartetsor jazz groups, whose harmonies arise in a self-organised way through mutual interactions. Thisis why the answer to the chicken and eggquestion in the origin of life is not that life beganwith DNA and RNA but that it must have begunwith primitive cells which provided theenvironment within which nucleic acids could besynthesised and serve as copying templates.

The paradox of development

It is a commonplace that despite the nearidentity of our genes, no one would mistake ahuman for a chimpanzee. For that matter we sharesome 35% of our genes with daffodils. What dis-tinguishes even closely related species are thedevelopmental processes that build on the genes,the ontogenetic mechanisms that transform thesingle fused cell of a fertilised ovum into thethousand trillion cells of the human body,hierarchically and functionally organised intotissues and organs. Developmental processeshave trajectories which constitute the individuallifeline of any organism, trajectories which areneither instructed by the genes, nor selected bythe environment, but constructed by theorganism out of the raw materials provided byboth genes and environment. This is theprocess described by Maturana & Varela16 asautopoiesis and by Oyama17 as “the ontogeny ofinformation”.

One of the problems for 20th century biology, aproblem resulting partly from contingent featuresof the history of our science, is that whereas at thebeginning of the century developmental biologyand genetics were seen as a single scientificendeavour (a splendid exemplar being T. H.Morgan, the founder of the famous “fly school”which introduced Drosophila as “god’s organism”for genetic and chromosomal analysis), by the1930s they had become quite distinct. Thusdevelopment became the science of similarities,genetics the science of differences. Explaining howit is that virtually all humans are between 1.5and 2.5 m in height, are more or less bilaterally


symmetrical and have pentadactyl limbs was asubject for developmental study. Why some ofus have differently coloured eyes, hair or skinbecame part of genetics. Only in the closing yearsof the century does there seem to be a chance tobring the two together once more.18

The unity of an organism is a process unity, not astructural one. All its molecules, and virtually allits cells, are continuously being transformed in acycle of life and death which goes on from themoment of conception until the final death of theorganism as a whole. This means that livingsystems are open, never in thermodynamic equi-librium and constantly choosing, absorbing andtransforming their environment. They are in con-stant flux, always at the same time both being andbecoming. To build on an example I owe to PatBateson, a newborn infant has a suckling reflex;within a matter of months the developing infantbegins to chew her food. Chewing is not simply amodified form of suckling, but involves differentsets of muscles and physiological mechanisms.The paradox of development is that a baby has tobe at the same time a competent suckler, and totransform herself into a competent chewer. To be,therefore, and to become.

Being and becoming are not to be partitionedinto that tired dichotomy of nature versus nurture.Rather they are defined by a different dichotomy,that of specificity and plasticity. The retina of theeye is connected via a series of neural stagingposts to the visual cortex at the back of the brain.A baby is born with most of these connections inplace, but during the first years of life, the eye andthe brain both grow, at different rates. This meansthat the connections between eye and brain havecontinually to be broken and remade. If thedeveloping child is to be able to retain a coherentvisual perception of the world this breaking andremaking must be orderly and relatively unmodifi-able by experience. This is specificity. However, asboth laboratory animal experiments and our ownhuman experience show, both the fine details ofthe “wiring” of the visual cortex, and how andwhat we perceive of the world are both directlyand subtly shaped by early experience. This isplasticity. All living organisms and perhapsespecially humans with our large brains showboth specificity and plasticity in development, andboth properties are enabled by our genes andshaped by our experience and contingency. Neithergenes nor environment are in this sense determi-nant of normal development; they are the rawmaterials out of which we autopoietically constructourselves.

Thus the four dimensions of living processes—three of space and one of time—cannot be read offfrom the one dimensional strand of DNA. A livingorganism is an active player in its own destiny,not a lumbering robot responding to geneticimperatives whilst passively waiting to discoverwhether it has passed what Darwin described asthe continuous scrutiny of natural selection.

Evolution and levels of selection

Within the reductionist paradigm in whichmuch of contemporary biological theorising islocated, the processes of evolutionary change havealso become simplified. By contrast for instancewith Darwin’s own pluralism, which saw naturalselection as a main but not the only mechanism ofchange, a dominant orthodoxy, described as funda-mentalist or ultra-Darwinian has emerged, asevidence in the popular writings of Dawkins.Three main theses characterise this new funda-mentalism. First, most phenotypic features wecan observe are adaptive; second, they are gener-ated by natural selection; and third, natural selec-tion acts solely or primarily at the level ofindividual genes. A new biology for the newmillennium must transcend each of thesepropositions, in part by reverting to Darwin’s ownmore pluralistic understanding. I will consider thecounter-arguments to these propositions in reverseorder.

The present-day understanding of the fluidgenome in which segments of DNA responsiblefor coding for subsections of proteins, or for regu-lating these gene functions, are distributed acrossmany regions of the chromosomes in which theDNA is embedded, and are not fixed in any onelocation but may be mobile, makes the view thatindividual genes are the only level of selectionuntenable. But it always was. To play their part inthe creation of a functioning organism many genesare involved. For us humans, the estimates havesteadily been revised down, to now some 30,000–40,000 genes. For an organism to survive and repli-cate, the genes are required to work in concert—that is, to cooperate. Antelopes which can outrunlions are more likely to survive and breed thanthose that cannot. Therefore a mutation in a genewhich improves muscle efficacy, for instance,might be regarded as fitter and therefore likely tospread in the population. However, as enhancedmuscle use requires other physiological adap-tations—such as increased blood flow to themuscles, without this concerted change in othergenes, the individual mutation is scarcely likely toprove very advantageous. And as many geneshave multiple phenotypic effects (pleiotropy) thelikelihood of a unidirectional phenotypic changeis complex—increased muscle efficacy mightdiminish the longevity of the heart for example.Thus it is not just single genes which get selected,but also genomes. Selection operates at the level ofgene, genome and organism.

Nor does it stop there. Organisms exist inpopulations (groups, demes). Three decades ago,Wynne-Edwards19 argued that selection occurredat the level of the group as well as the individual.He based this claim on a study of a breedingpopulation of red grouse on Scottish moors, andargued that they distributed themselves across themoor, and regulated their breeding practices, in away which was optimum for the group as a whole


rather than any individual member within it. Itmay be in the individual’s interest to produce lotsof offspring; but this might overcrowd the moor,which could only sustain a smaller number ofbirds; hence it is in the group’s interest that noneof its members over-breed. Orthodox Darwinians,led by George Williams, treated this claim with asmuch derision as they did Lamarck’s view thatacquired characteristics could be inherited, andgroup selection disappeared from the literature.

Today however it is clear that the attack wasmisjudged. In part it was semantic. MaynardSmith’s own work indicated that stable popu-lations require the mutual interactions and ratiosof members with very different types ofbehaviours (hawks and doves, for instance, to usehis model)—so called evolutionary stablestrategies. But there are an increasing number ofexamples of populations of organisms whosebehaviours can most economically be described bygroup selectionist equations. Recently Sober &Wilson20 have published a major reassessment ofgroup selectionist models and shown mathe-matically how even such famously counter-intuitive (for ultra-Darwinians) phenomena asaltruism can occur, in which an individual sacri-fices its own individual fitness, not merely for theinclusive fitness of its kin but for the benefit of thegroup as a whole.

Finally, there is selection at even higher levels—that of the species for example.21 Natural selectionmay be constantly scrutinising and honing theadaptiveness of a particular species to its environ-ment, but cannot predict the consequences ofdramatic changes in that environment, as forexample the meteor crash into the Yucatanbelieved to have precipitated the demise of thedinosaurs. Selection also operates at the level ofentire ecosystems. Consider, for example, a beaverdam. Dawkins22 uses this example to claim thatthe dam may be regarded as part of the beaver’sphenotype—thus swallowing an entire smalluniverse into the single strand of DNA. But if it isa phenotype, it is the phenotype of many beaversworking in concert, and indeed of the many com-mensal and symbiotic organisms which also liveon and modify for their needs the structure of thedam. As Sober and Wilson point out, selectionmay indeed occur at the level of the individual,but what constitutes an individual is very much inthe eye of the beholder. Genes are distributedacross genomes within an organism, and theyare also distributed across groups of organismswithin a population. There is no overriding reasonwhy we should consider “the organism” as anindividual rather than “the group” or even “theecosystem”.

Nor is natural selection the only mode of evol-utionary change. We need not be Lamarckian toaccept that other processes are at work. Sexualselection is one well-accepted mechanism. Theexistence of neutral mutations, founder effects,genetic drift, and molecular drive,23 all enrich the

picture. Gould24 argues that much evolutionarychange is contingent, accidental, and that, as heputs it, if one were to wind the tape of historybackward and replay it, it is in the highest degreeunlikely that mammals, let alone humans wouldevolve.

Finally, not all phenotypic characters are adap-tive. A core assumption of ultra-Darwinism is thatobserved characters must be adaptive, so as to pro-vide the phenotypic material upon which naturalselection can act. However, what constitutes a char-acter—and what constitutes an adaptation—is asmuch in the eye of the beholder as in the organismto which the “character” belongs. The problem liesin part in the ambiguity of the term phenotypewhich can refer to anything from a piece of DNA(strictly the gene’s phenotype) through the cellularexpression of a protein to a property of the organ-ism as a whole, like height, or “behaviour” suchas gait. At which level of phenotype a character is“adaptive”, if at all, and at which its propertiesare epiphenomenal is always going to be a matterfor debate. A not entirely apocryphal example isprovided by the American artist Thayer, whosuggested, early in this century, that the flamingo’spink coloration is an adaptation to make them lessvisible to predators against the pink evening sky.But the coloration is a consequence of theflamingo’s shrimp diet and fades if the dietchanges. Thus even if we were to assume that thecoloration was indeed protective, it is an epipheno-menal consequence of a physiological—dietary—adaptation, rather than a selected property in itsown right.

Natural selection’s continual scrutiny does notgive it an a la carte freedom to accept or reject geno-typic or phenotypic variation. Structural con-straints insist that evolutionary, geneticmechanisms are not infinitely flexible but mustwork within the limits of what is physically orchemically possible. For instance, the limits to thesize of a single cell are set by the physics of diffu-sion processes, the size of a crustacean like alobster or crab by the constraints of its exoskeleton.The limits to the possible lift of any conceivablewing structure make it impossible to geneticallyengineering humans to sprout wings and fly; thereare good reasons why we cannot become angels.Webster & Goodwin25 have extended this argu-ment further, arguing that there are exact “laws ofform” which ensure, for instance the generation ofpentadactyl (five-fingered) limbs.

A biological decalogue for the millennium

If we are to transcend biology’s reductionistview of nature for the new millennium, and tocreate what I would regard as an understandingof living processes more in accord with thematerial reality of the world than our present,rather one-eyed view, we need some principleswith which to work. These principles will of coursenot reject the explanatory power of reductionism,


but will recognise its limitations. They shouldaccept too, Goodwin’s26 call for the return to ascience of qualities, to complement reductionistquantitation. Such a science will rejoice in com-plexity, in dynamics, and in an emphasis as muchon process as on objects. In Lifelines,1 I enumeratea set of such principles; by chance rather thandesign, there are ten in all. In brief, here is mydecalogue. To exemplify all its principles wouldextend this essay beyond reasonable length, butenough has I trust been said to give a feel for theconceptual approach I am arguing for:

(1) Scientific knowledge is not absolute, butprovisional, being socially, culturally,technologically and historically constrained.

(2) We live in a world that is ontologicallyunitary, but our knowledge of it is epistemo-logically diverse. There are multiple legiti-mate ways of describing and explainingany living process.

(3) Different sciences deal with different levelsof organisation of matter of increasing com-plexity. Terms and concepts applicable atone level are not necessarily applicable atothers. Thus genes cannot be selfish; it ispeople, not neurons nor yet brains or mind,who think, remember and show emotion.

(4) Causes are multiple, and phenomena richlyinterconnected. Adequate explanationdemands finding the determining level. Totake an example, the high levels of murderin the US by comparison with say Europeor Japan are best explained not by somespecial feature of the US genotype, abnor-mal genes or biochemistry which predisposeto violence (despite a major research pro-gramme dedicated to identifying such bio-logical predispositions), but by the highnumber of personal handguns in society,and a culture and history of their use.

(5) Living organisms exist in four dimensions,three of space and one of time, a develop-mental trajectory or lifeline, always auto-poietic, both being and becoming. Lifelinesare stabilised through dynamic processes.The traditional biological concept of homeo-stasis as a regulatory mechanism needstransforming by that of homeodynamics, toemphasise that indeed stasis for any livingorganism means death.

(6) Organism and environment interpenetrate;environments select organisms (the pro-cess of natural selection); but organismschoose and transform environments.Organisms are thus active players in theirown destiny.

(7) Living organisms are open systems far fromthermodynamic equilibrium; continuity ismaintained by a constant flow of energyand information. All is flux; stabilityemerges through process, not structure.

(8) Evolutionary change occurs at the inter-

section of lifeline trajectories with changingenvironments.

(9) Organisms cannot predict patterns ofchange, and selection therefore alwaystracks environmental change. Nothing inbiology makes sense except in the contextof history.

(10) Thus the future, for humans and otherliving organisms, is radically unpredictable;we make our own history, though in circum-stances not of our own choosing.

References

1. Rose, S. P. R. (1997). Lifelines: Biology, Freedom,Determinism, Allen Lane The Penguin Press, London.

2. Nagel, T. (1998). Reductionism and antireductionism.Novartis Symposium: The Limits of Reductionism inBiology, pp. 3–14, Wiley, Chichester.

3. Wilson, E. O. (1975). Sociobiology: the New Synthesis,Harvard University Press, Cambridge.

4. Wilson, E. O. (1998). Consilience: The Unity ofKnowledge, Little Brown, London.

5. Midgley, M. (1998). One world, but a big one. InBrains to Consciousness? Essays on the New Sciences ofthe Mind (Rose, S., ed.), pp. 246–270, Allen Lane ThePenguin Press, London.

6. Churchland, P. (1986). Neurophilosophy: Towards aUnified Science of the Mind/Brain, MIT Press,Cambridge, MA.

7. Dennett, D. (1991). Consciousness Explained, LittleBrown, Boston.

8. Dawkins, R. (1976). The Selfish Gene, OxfordUniversity Press, Oxford.

9. Rose, H. A. (1994). Love, Power and Knowledge, PolityPress, Oxford.

10. Barkow, J. H., Cosmides, L. & Tooby, J. (eds) (1992).The Adapted Mind: Evolutionary Psychology and theGeneration of Culture, Oxford University Press, NewYork.

11. Rose, H. A. & Rose, S. P. R. (eds) (2000). Alas PoorDarwin: Arguments against Evolutionary Psychology,Cape, London.

12. Freeman, W. (1999). How Brains Make up Their Minds,Weidenfeld and Nicolson, London.

13. Pinker, S. (1997). How the Mind Works, Allen LaneThe Penguin Press, London.

14. Ellman, J. L., Bates, E. A., Johnson, M. H., Karmiloff-Smith, A., Parisi, D. & Plunkett, K. (1996). RethinkingInnateness: A Connectionist Perspective on Development,MIT Press, Cambridge, MA.

15. Kauffman, S. (1995). At Home in the Universe: theSearch for Laws of Complexity, Allen Lane The PenguinPress, London.

16. Maturana, H. R. & Varela, F. J. (1998). The Tree ofKnowledge: The Biological Roots of Human Understand-ing, Shambhala Press, Boston.

17. Oyama, S. (1985). The Ontogeny of Information,Cambridge University Press, Cambridge.

18. Maynard Smith, J. (1998). Shaping Life: Genes, Embryosand Organisms, Weidenfeld and Nicolson, London.

19. Wynne-Edwards, (1962). Animal Dispersion in Relationto Social Behaviour, Oliver and Boyd, Edinburgh.

20. Sober, E. & Wilson, D. S. (1998). Unto Others; theEvolution and Psychology of Unselfish Behavior,Harvard University Press, Cambridge, MA.


21. Gould, S. J. (2002). The Structure of EvolutionaryTheory, Harvard University Press, Cambridge, MA.

22. Dawkins, R. (1982). The Extended Phenoptype: The Geneas the Unit of Selection, Freeman, Oxford.

23. Dover, G. (2000). Dear Mr Darwin: Letters on theEvolution of Life and Human Nature, Weidenfeld andNicolson, London.

24. Gould, S. J. (1989). Wonderful Life: The Burgess Shaleand the Nature of History, Penguin, Harmondsworth.

25. Webster, G. & Goodwin, B. (1996). Form and Trans-formation: Generative and Relational Principles inBiology, Cambridge University Press, Cambridge.

26. Goodwin, B. (1994). How the Leopard Changed itsSpots, Weidenfeld and Nicolson, London.

Steven Rose has been Professor of Biology and Director of the Brain and Behaviour Research Group at The OpenUniversity, Milton Keynes in the UK since the establishment of the university in 1969, His research is focussed onunderstanding the cellular and molecular mechanisms of learning and memory on which he has published some 300research papers. His international honours include the Sechenov and Anokhin Medals (Russia), the Ariens Kappersmedal (The Netherlands) and the Biochemical Society medal for public communication of science. As well as hisposition at the Open University he is currently also joint Professor of Physic at Gresham College London (a post heldjointly with Professor Hilary Rose, with a remit to lecture on “genetics and society”). He is also a Visiting Professorin the Department of Anatomy and Developmental Biology at University College London. As well as his researchpapers in neuroscience and related fields Steven Rose has written or edited 15 books including The Chemistry of Life,Science and Society (with Hilary Rose); The Conscious Brain, Not in our Genes (with Richard Lewontin and Leo Kamin),The Making of Memory winner of the 1993 Rhone-Poulenc/Royal Society Science Book Prize, Lifelines, and most recently,edited jointly with Hilary Rose Alas Poor Darwin: Arguments Against Evolutionary Psychology. He is currently working ona book on the future of the brain.

(Received and Accepted 11 April 2002)


the biology of the future and the future of biology

Documents