AbstractBehavior genetics has dem-

onstrated that genetic varianceis an important component ofvariation for all behavioral out-comes, but variation amongfamilies is not. These resultshave led some critics of behav-ior genetics to conclude thatheritability is so ubiquitous asto have few consequences forscientific understanding of de-velopment, while some be-havior genetic partisans haveconcluded that family environ-ment is not an important causeof developmental outcomes.Both views are incorrect. Geno-type is in fact a more system-atic source of variability thanenvironment, but for reasonsthat are methodological ratherthan substantive. Developmentis fundamentally nonlinear,interactive, and difficult to con-trol experimentally. Twin stud-ies offer a useful methodologi-cal shortcut, but do not showthat genes are more fundamen-tal than environments.

Keywordsgenes; environment; develop-ment; behavior genetics

The nature-nurture debate isover. The bottom line is that every-thing is heritable, an outcome thathas taken all sides of the nature-nurture debate by surprise. IrvingGottesman and I have suggestedthat the universal influence ofgenes on behavior be enshrined asthe first law of behavior genetics(Turkheimer & Gottesman, 1991),and at the risk of naming laws thatI can take no credit for discovering,it is worth stating the nearly unani-mous results of behavior geneticsin a more formal manner.

● First Law. All human behavioraltraits are heritable.

● Second Law. The effect of beingraised in the same family issmaller than the effect of genes.

● Third Law. A substantial portionof the variation in complex hu-man behavioral traits is not ac-counted for by the effects ofgenes or families.

It is not my purpose in this briefarticle to defend these three lawsagainst the many exceptions thatmight be claimed. The point is thatnow that the empirical facts are inand no longer a matter of seriouscontroversy, it is time to turn atten-tion to what the three laws mean,

to the implications of the geneticsof behavior for an understandingof complex human behavior and itsdevelopment.

VARIANCE ANDCAUSATION INBEHAVIORAL

DEVELOPMENT

If the first two laws are taken lit-erally, they seem to herald a greatvictory for the nature side of theold debate: Genes matter, familiesdo not. To understand why suchviews are at best an oversimplifica-tion of a complex reality, it is nec-essary to consider the newest waveof opposition that behavior genet-ics has generated. These new crit-ics, whose most articulate spokes-man is Gilbert Gottlieb (1991, 1992,1995), claim that the goal of devel-opmental psychology is to specifythe actual developmental processesthat lead to complex outcomes. Inlower animals, whose breedingand environment can be broughtunder the control of the scientist, itis possible to document such devel-opmental processes in exquisite de-tail. The critics draw an unfavor-able comparison between thesedetailed animal studies and twinstudies of behavior genetics, whichproduce only statistical conclu-sions about the relative importanceof genes and environment in devel-opment.

The greatest virtue of the newchallenge is that it abandons the

Three Laws of Behavior Genetics andWhat They MeanEric Turkheimer1

Department of Psychology, University of Virginia, Charlottesville, Virginia

160 VOLUME 9, NUMBER 5, OCTOBER 2000

Published by Blackwell Publishers Inc.

implausible environmentalist con-tention that important aspects ofbehavior will be without geneticinfluence. Gottlieb (1992) stated,“The present . . . viewpoint holdsthat genes are an inextricable com-ponent of any developmental sys-tem, and thus genes are involved inall traits” (p. 147). Unlike earliercritics who deplored the reduction-ism they attributed to behaviorgenetic theories of behavior, thedevelopmental biologists take be-havior genetics to task for not be-ing mechanistic enough. Once vili-fied as the paragon of deterministaccounts of human behavior, be-havior genetics is now chastised foroffering vague and inconclusivemodels of development (Gottlieb,1995; Turkheimer, Goldsmith, &Gottesman, 1995), and judged bythe standards of developmentalpsychobiology in lower animals, it

is true enough that behavior ge-netic theories of complex humanbehavior seem woefully poorlyspecified. But ultimately the chargeis unfair, because there is noequivalent in developmental psy-chobiology to the behavior geneticstudy of marital status or schoolperformance. The great preponder-ance of the exquisite experimentalscience that goes into animal psy-chobiology is quite simply impos-sible to conduct in humans.

Human developmental socialscience is difficult—equally so forthe genetically and environmen-tally inclined—because of the(methodologically vexing, human-istically pleasing) confluence oftwo conditions: (a) Behavioremerges out of complex, nonlineardevelopmental processes, and (b)ethical considerations prevent usfrom bringing most human de-

velopmental processes under effec-tive experimental control. Figure 1is a schematic illustration of theproblem. Individual genes (Genes1, 2, and 3) and their environments(which include other genes) inter-act to initiate a complex develop-mental process that determinesadult personality. Most characteris-tic of this process is its interactivity:Subsequent environments to whichthe organism is exposed depend onits earlier states, and each new en-vironment changes the develop-mental trajectory, which affects fu-ture expression of genes, and soforth. Everything is interactive, inthe sense that no arrows proceeduninterrupted from cause to effect;any individual gene or environ-mental event produces an effectonly by interacting with othergenes and environments.

For the behavior geneticist,

Fig. 1. Schematic diagram of contrasting roles of genes and environment in development of personality. One-headed arrows linkcauses to effects; two-headed arrows indicate correlations. Genes and environments are both causal inputs into an interactivedevelopmental system (represented by the network of arrows in the center of the figure), but because people select and shape theirown environments (as represented by lighter one-headed arrows from personality to environments), correlations across thedevelopmental system (dotted two-headed arrows) are easier to detect for genes than for environments.

161CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE

Copyright © 2000 American Psychological Society

however, the quasi-experimentalgift of genetically identical andnonidentical twins offers a remark-able, if deceptively simple, methodto span this daunting interactivecomplexity. Thanks to the fact thatidentical twins are on average ex-actly twice as similar genetically asnonidentical twins, one can usestraightforward statistical proce-dures to estimate the proportion ofvariability in complex outcomesthat is associated with causally dis-tant genes, all the while maintain-ing a state of near-perfect igno-rance about the actual causalprocesses that connect genes to be-havior. This methodological short-cut is not available to rivals of be-havior genetics who seek tomeasure the effects of families onbehavior. How similar was myrearing environment to that of mysiblings? And how similar was it tothe environment of my adoptedsibling, if I have one, or to the en-vironment of my biological siblingwho was raised by someone else?The apparent victory of natureover nurture suggested by the firsttwo laws is thus seen to be moremethodological than substantive.In a world in which there were oc-casional occurrences of “identicalenvironmental twins,” whose ex-periences were exactly the same,moment by moment, and anothervariety who shared exactly (butrandomly) 50% of their experi-ences, environmentalists could re-produce the precision of their ri-vals , and l ike the behaviorgeneticists could measure withgreat precision the total contribu-tion of the environment whileknowing almost nothing about thedevelopmental processes that un-derlie it.

The old-fashioned nature-nur-ture debate was about whether ornot genes influence complex be-havioral outcomes, and that ques-tion has been decisively answeredin the affirmative. The new ques-tion is how we can proceed from

partitioning sources of variance tospecifying concrete developmentalprocesses (Turkheimer, 1998), andalthough critics like Gottlieb arecorrect that heritability per se hasfew implications for a scientific un-derstanding of development, theyhave failed to emphasize two cru-cial points. First, heritability doeshave one certain consequence: It isno longer possible to interpret cor-relations among biologically re-lated family members as prima fa-cie evidence of sociocultural causalmechanisms. If the children of de-pressed mothers grow up to be de-pressed themselves, it does notnecessarily demonstrate that beingraised by a depressed mother is it-self depressing. The children mighthave grown up equally depressedif they had been adopted andraised by different mothers, underthe influence of their biologicalmother’s genes. For every behaviorgeneticist who continues to reportmoderate heritabilities as thoughthey were news, there is an envi-ronmentalist who reports causallyambiguous correlations betweengenetically related parents andchildren. Second, the problem thecritics have uncovered extendswell beyond behavior genetics: It isa rare environmentalist who hasnever used statistical methods topredict behavioral outcomes fromearlier events, in the hope that thespecific developmental mecha-nisms can be filled in later. The dis-connect between the analysis ofvariance and the analysis of causes,to use Lewontin’s (1974) phrase, isnot a proprietary flaw in behaviorgenetic methodology; in fact, it isthe bedrock methodological prob-lem of contemporary social science.

NONSHAREDENVIRONMENT AND THE

GLOOMY PROSPECT

Even after the effects of genesand the shared effects of families

have been accounted for, around50% of the differences among sib-lings is left unexplained. In recentyears, scientists interested in thegenetics of behavior have come tocall this unexplained portion the“nonshared environment.” Al-though according to the secondlaw shared environment accountsfor a small proportion of the vari-ability in behavioral outcomes, ac-cording to the third law, nonsharedenvironment usually accounts for asubstantial portion. So perhaps theappropriate conclusion is not somuch that the family environmentdoes not matter for development,but rather that the part of the fam-ily environment that is shared bysiblings does not matter. Whatdoes matter is the individual envi-ronments of children, their peers,and the aspects of their parentingthat they do not share. Plomin andDaniels (1987) reviewed evidenceof the predominance of nonsharedenvironmental variance and poseda seminal question: Why are chil-dren in the same family so differ-ent? They proposed that siblingsare different because nonsharedenvironmental events are morepotent causes of developmentaloutcomes than the shared environ-mental variables, like socioeco-nomic status, that have formed thetraditional basis of socioculturaldevelopmental psychology.

Plomin and Daniels’s explana-tion involves a subtle conceptualshift, best described in terms of adistinction between the objectiveand effective environment (Gold-smith, 1993; Turkheimer & Wal-dron, 2000). What qualifies an en-vironmental event as nonshared?There are two possibilities. Thefirst is objective: An event is non-shared if it is experienced by onlyone sibling in a family, regardlessof the consequences it produces.The other possibility is effective:An environmental event is non-shared if it makes siblings different

162 VOLUME 9, NUMBER 5, OCTOBER 2000

Published by Blackwell Publishers Inc.

rather than similar, regardless ofwhether it was experienced by oneor both of them. Plomin andDaniels’s proposal, then, is that thenonshared environment as an ef-fectively defined variance compo-nent can be explained by objec-tively nonshared environmentalevents. The question, “Why arechildren in the same family so dif-ferent?” is answered, “Becausemeasurable differences in their en-vironments make them that way.”

This proposal has been enor-mously influential, spawning anentire area of empirical inquiry intothe consequences of measured en-vironmental differences amongsiblings. Ironically, that same lit-erature has quite decisively dem-onstrated that the conjecture isfalse. A review of 43 studies thatmeasured differences in the envi-ronments of siblings and relatedthem to differences in the siblings’developmental outcomes (Turk-heimer & Waldron, 2000) hasshown that although upwards of50% of the variance in behavioraloutcomes is accounted for by theeffectively defined variance com-ponent called nonshared environ-ment, the median percentage ac-counted for by objectively definednonshared events is less than 2%.What could be going on?

Plomin and Daniels (1987) al-most identified the answer to thisquestion, but dismissed it as toopessimistic:

One gloomy prospect is that the salientenvironment might be unsystematic,idiosyncratic, or serendipitous eventssuch as accidents, illnesses, or othertraumas . . . . Such capricious events,however, are likely to prove a dead endfor research. More interesting heuristi-cally are possible systematic sources ofdifferences between families. (p. 8)

The gloomy prospect is true. Non-shared environmental variabilitypredominates not because of thesystematic effects of environmental

events that are not shared amongsiblings, but rather because of theunsystematic effects of all environ-mental events, compounded by theequally unsystematic processesthat expose us to environmentalevents in the first place (Turk-heimer & Gottesman, 1996).

A model of nonshared variabil-ity based on the gloomy prospect isradically different from the Plominmodel based on systematic conse-quences of environmental differ-ences among siblings. Most impor-tant, the two models suggest verydifferent prospects for a geneticallyinformed developmental psychol-ogy. Again and again, Plomin andhis colleagues have emphasizedthat the importance of nonsharedenvironment implies that it is timeto abandon shared environmentalvariables as possible explanationsof developmental outcomes. Andalthough modern environmental-ists might not miss coarse mea-sures like socioeconomic status, itis quite another thing to give up onthe causal efficaciousness of nor-mal families, as Scarr (1992), Rowe(1994), and Harris (1998) haveurged. If, however, nonshared en-vironmental variability in outcomeis the result of the unsystematicconsequences of both shared andnonshared environmental events,the field faces formidable method-ological problems—Plomin andDaniels’s gloomy prospect—butneed not conclude that aspects offamilies children share with sib-lings are of no causal importance.

CONCLUSION:ANTICIPATING THEGENOME PROJECT

It is now possible for behaviorgenetics to move beyond statisticalanalyses of differences betweenidentical and nonidentical twinsand identify individual genes thatare related to behavioral outcomes.

What should we expect from thisendeavor? Behavior geneticists an-ticipate vindication: At long last,statistical variance componentswill be rooted in the actual causalconsequences of actual genes. Crit-ics of behavior genetics expect theopposite, pointing to the repeatedfailures to replicate associations be-tween genes and behavior as evi-dence of the shaky theoretical un-derpinnings of which they have solong complained.

There is an interesting parallelbetween the search for individualgenes that influence behavior andthe failed attempt to specify thenonshared environment in terms ofmeasured environmental variables.In each case, investigators beganwith statistically reliable but caus-ally vague sources of variance, andset out to discover the actual causalprocesses that produced them. Thequest for the nonshared environ-ment, as we have seen, got stuck inthe gloomy prospect. Although in-dividual environmental events in-fluence outcomes in the most gen-eral sense, they do not do so in asystematic way. One can detecttheir effects only by accumulatingthem statistically, using twins oradoptees.

If the underlying causal struc-ture of human development ishighly complex, as illustrated inFigure 1, the relatively simple sta-tistical procedures employed bydevelopmental psychologists, ge-neticists, and environmentalistsalike are being badly misapplied.But misapplied statistical proce-dures still produce what appear tobe results. Small relations wouldstill be found between predictorsand outcomes, but the underlyingcomplex causal processes wouldcause the apparent results to besmall, and to change unpredictablyfrom one experiment to the next. Soindividual investigators would ob-tain “results,” which would thenfail to replicate and accumulateinto a coherent theory because the

163CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE

Copyright © 2000 American Psychological Society

simple statistical model did not fitthe complex developmental pro-cess to which it was being applied.Much social science conducted inthe shadow of the gloomy prospecthas exactly this flavor (e.g., Meehl,1978).

The gloomy prospect loomslarger for the genome project thanis generally acknowledged. Thequestion is not whether there arecorrelations to be found betweenindividual genes and complex be-havior—of course there are—butinstead whether there are domainsof genetic causation in which thegloomy prospect does not prevail,allowing the little bits of correla-tional evidence to cohere into rep-licable and cumulative geneticmodels of development. My ownprediction is that such domainswill prove rare indeed, and that thelikelihood of discovering them willbe inversely related to the com-plexity of the behavior understudy.

Finally, it must be rememberedthat the gloomy prospect is gloomyonly from the point of view of theworking social scientist. Althoughfrustrated developmental psy-chologists may be tempted to favormethodologically tractable heuris-

tics over chaotic psychological re-ality, it is a devil’s choice: In thelong run, the gloomy prospect al-ways wins, and no one would wantto live in a world where it did not.Psychology is at least one goodparadigm shift away from an em-pirical answer to the gloomy pros-pect, but the philosophical re-sponse is becoming clear: Theadditive effect of genes may con-stitute what is predictable abouthuman development, but what ispredictable about human develop-ment is also what is least interest-ing about it. The gloomy prospectisn’t.

Recommended Reading

Gottlieb, G. (1992). (See References)Lewontin, R.C. (1974). (See Refer-

ences)Meehl, P.E. (1978). (See References)Plomin, R., & Daniels, D. (1987). (See

References)

Note

1. Address correspondence to EricTurkheimer, Department of Psychol-ogy, 102 Gilmer Hall, P.O. Box 400400,University of Virginia, Charlottesville,VA 22904-4400; e-mail: turkheimer@virginia.edu.

ReferencesGoldsmith, H. (1993). Nature-nurture issues in the

behavioral genetic context: Overcoming barri-ers to communication. In R. Plomin & G. Mc-Clearn (Eds.), Nature, nurture and psychology(pp. 325–339). Washington, DC: AmericanPsychological Association.

Gottlieb, G. (1991). Experiential canalization of be-havioral development: Theory. DevelopmentalPsychology, 27, 4–13.

Gottlieb, G. (1992). Individual development and evo-lution. New York: Oxford University Press.

Gottlieb, G. (1995). Some conceptual deficienciesin “developmental” behavior genetics. HumanDevelopment, 38, 131–141.

Harris, J.R. (1998). The nurture assumption: Whychildren turn out the way they do. New York:Free Press.

Lewontin, R.C. (1974). The analysis of varianceand the analysis of causes. American Journal ofHuman Genetics, 26, 400–411.

Meehl, P.E. (1978). Theoretical risks and tabularasterisks: Sir Karl, Sir Ronald, and the slowprogress of soft psychology. Journal of Consult-ing and Clinical Psychology, 46, 806–834.

Plomin, R., & Daniels, D. (1987). Why are childrenin the same family so different from one an-other? Behavioral and Brain Sciences, 10, 1–60.

Rowe, D.C. (1994). The limits of family influence:Genes, experience, and behavior. New York: Guil-ford Press.

Scarr, S. (1992). Developmental theories for the1990s: Development and individual differ-ences. Child Development, 63, 1–19.

Turkheimer, E. (1998). Heritability and biologicalexplanation. Psychological Review, 105, 782–791.

Turkheimer, E., Goldsmith, H.H., & Gottesman,I.I. (1995). Commentary. Human Development,38, 142–153.

Turkheimer, E., & Gottesman, I.I. (1991). Is H2 = 0a null hypothesis anymore? Behavioral andBrain Sciences, 14, 410–411.

Turkheimer, E., & Gottesman, I.I. (1996). Simulat-ing the dynamics of genes and environment indevelopment. Development and Psychopathol-ogy, 8, 667–677.

Turkheimer, E., & Waldron, M.C. (2000). Non-shared environment: A theoretical, method-ological, and quantitative review. PsychologicalBulletin, 126, 78–108.

164 VOLUME 9, NUMBER 5, OCTOBER 2000

Published by Blackwell Publishers Inc.