This article is available at http://www.nhm.org/research/annelida/eb_vs_id.html1

email: kfitzhug@nhm.org2

-1-

EVOLUTIONARY BIOLOGY VERSUS INTELLIGENT DESIGN:RESOLVING THE ISSUE1

J. Kirk Fitzhugh, Ph.D.2

Research & CollectionsNatural History Museum of Los Angeles County

-2-

INTRODUCTION TO THE PROBLEM

In recent years, the public has witnessed agreat deal of attention being given to thesubject of teaching evolutionary biology in

public school systems. Most notable has beenthe view that students should be presented notonly with what is often termed ‘the theory ofevolution,’ but that this theory should beconsidered alongside what has become knownas ‘intelligent design’ (ID). Part of the basisfor the reasoning behind this move has beenthe often-stated phrase, ‘Evolution is just atheory, not a fact,’ and ID can be treated in ascientific context just as readily as the theoryof evolution. For instance, in his book,Intelligent Design (1999, pg. 13), W.A.Dembski states, ‘Intelligent design is threethings: a scientific research program thatinvestigates the effects of intelligent cause; anintellectual movement that challengesDarwinism and its naturalistic legacy; and away of understanding divine action.’ As Iwill discuss in this article, the reason ID isseen as a viable scientific alternative can betraced largely to misunderstandings of theterms ‘theory,’ ‘hypothesis,’ and ‘fact,’ aswell as the process of critically evaluatingtheories and hypotheses routinely applied inall fields of science. I address thesemisunderstandings for the purpose ofproviding the lay person with a better idea ofwhat these terms mean and how they are usedin science, as well as in everyday life. Thegoal is not to address the ‘debate’ oftenpresented in the media under the guise of‘evolution versus ID,’ as there is no actualscientific debate. Rather, I wish to show thatthe foundations for teaching science, asopposed to non-scientific approaches, requiresaccurate renderings of the principles thatallow for the rational acquisition ofunderstanding.

THEORIES, FACTS, AND SCIENCE:MYTHS AND MISCONCEPTIONS

In 2004, the school board for Dover,Pennsylvania, voted to have ninth-gradebiology teachers read to students thefollowing statement:

The Pennsylvania Academic Standardsrequire students to learn about Darwin'sTheory of Evolution and eventually to take astandardized test of which evolution is a part.Because Darwin's Theory is a theory, itcontinues to be tested as new evidence isdiscovered. The Theory is not a fact. Gaps inthe Theory exist for which there is noevidence. A theory is defined as a well-testedexplanation that unifies a broad range ofobservations. Intelligent Design is anexplanation of the origin of life that differsfrom Darwin's view. The reference book, OfPandas and People, is available for studentswho might be interested in gaining anunderstanding of what Intelligent Designactually involves. With respect to any theory,students are encouraged to keep an openmind. The school leaves the discussion of theOrigins of Life to individual students and theirfamilies.

Notice that the Dover statement makesreference to a popular point of view, that‘evolution is a theory, not a fact.’ This has theintended effect of implying that theories areless credible than facts, and since evolution is‘just a theory,’ then it is reasonable toconsider ID as a viable alternative. Whilemuch of the recent discussion surrounding theteaching of evolutionary biology in publicschools has centered on views that ‘Darwin'sTheory is a theory,’ and, ‘The Theory is not afact,’ what these claims are intended to showis not correct. We need to understand what ismeant when we use terms like fact, theory,and hypothesis in the context of science in

-3-

order to engage in discussions regarding theutility of evolutionary biology relative to ID.

What are ‘facts?’Let’s consider a simple example to illustratewhat is meant when we use the term fact.Sitting in front of you on a table are theobjects shown in Figure 1.

What you observe are facts. All objectsand events you observe and experience, aswell as those you do not, are facts. Facts arethe objects and events that exist around us,and even within us.

Perceptions and hypotheses.In looking at the objects in Figure 1, youmight say that you ‘see’ some facts. But whatdoes this mean? The first reaction you haveto experiencing facts is that your brainproduces a belief, which in this instance is theresult of the rays of light hitting the retinas ofyour eyes, producing nerve impulses thattravel to your brain, producing your mentalreaction to the facts in front of you. There arerelations between the facts and you, the

perceiver, that causes you to believesomething is the case. Namely, that ‘there isa glass of ice water on the table.’ In otherwords, you have developed beliefs about thefacts in front of you.

Two interrelated beliefs might be that (1)the object does indeed exist in front of you, asopposed to being a hallucination, and (2) whatdoes exist in front of you is a glass of icewater. The process of reasoning from yoursenses to the belief that a glass of ice water ison the table is an attempt to explain the factsbefore you. In other words, you havedeveloped a hypothesis. A hypothesis is anexplanation of some set of facts, giving us atleast initial understanding of what weperceive.

Having your belief, that what exists infront of you is a glass of ice water, you mightthen proclaim to others, 'Here is a glass of icewater.' You have communicated yourhypothesis, perhaps for the purpose of givingothers understanding of what they haveobserved of the same facts.

How did you arrive at that hypothesis?What was the basis for reasoning that the factsbefore you, that you perceived via yoursenses, can be hypothesized to be a 'glass ofice water?' What allowed you to develop thatparticular hypothesis is that you applied a setof theories to what you perceived. At leasttwo theories are easy to recognize in thisinstance: a theory of glass and a theory ofwater. A theory is an established or generallyaccepted explanatory concept, or set ofconcepts, that we apply to our senseperceptions to give us understanding of whatwe do or might perceive. Theories are usuallystated as cause and effect relations. Thus, therelations between one’s observations, aseffects, and a theory would be to apply thattheory to observations to develop a specifichypothesis of cause that accounts for theobserved effects. Unlike hypotheses, whichFigure 1. The facts.

-4-

relate specific past causes to effects observedin the present, theories are time independent.They are applicable to the past, present, andfuture. For instance, we expect our theories ofglass and water would be useful throughoutthe universe, but the hypothesis ‘There is aglass of ice water on my table,’ only applies toparticular objects at a specific point in time.As we will see later, understanding thisdistinction is critical to the matter ofevaluating evolutionary and ID theories.

It is when we apply some theory to ourperceptions that we formulate beliefs andobservation statements, i.e., hypotheses. Forinstance, we regard the theories of glass andwater to be very well established given thatwe have had good success in the past inexplaining a variety of observations referredto as glass and water, and we have beensuccessful at making predictions based onthose theories. Our understanding of thenature of glass and water are so wellestablished that we do not hesitate to applytheories that represent that understanding.Thus, when we speak of our beliefs,observation statements, or hypotheses, these

are all products of the interplay between factsand particular theories in the process of tryingto understand our perceptions (Figure 2).

Hypotheses and theories are fallible.We concluded earlier that what we see is aglass of ice water. But, there are never anyguarantees that hypotheses, or the theoriesupon which they are based, are true. Forinstance, your friend might tell you that shethinks your hypothesis is not entirely correct,that the explanation of your senses is wrong,that at least some of the facts are not what youclaim. In this case, what she would besuggesting is that you did not apply therelevant theory to your sense data. She mightclaim that, rather than being filled with icewater, the glass contains lucite plastic made tolook like ice water. You might decide to thenapply another theory (or theories) to yoursenses, say a theory about lucite, and as aresult, you then consider a new hypothesisregarding the facts.

Notice that we are talking about thefallibility of hypotheses, as well as theories.While a particular hypothesis or theory might

Figure 2. The interplay between facts, observer, and theory to

produce a hypothesis.

-5-

be shown to be incorrect, the notions of truthand falsity do not apply to facts. Facts arewhat exist, regardless of who might be aroundto experience them, or not. It is you and I, asobservers, who can be incorrect in ourattempts to explain what we perceive.

Why be concerned with the nature of facts,hypotheses, and theories?Humans want to acquire understanding of theworld (and universe) around them. We havea remarkable capacity to perceive oursurroundings, but that is a minor part of whatwe do. We want to understand what weperceive. All fields of science hold this desirefor understanding as the ultimate goal. But,for this to be achieved, we must know whatwe mean when we use the words fact,hypothesis, and theory. The ‘glass of icewater’ example shows how we use theoriesand hypotheses to enable us to acquireunderstanding of the facts with which wecome into contact (Fig. 2).

So, what is wrong with the Dover PA schoolboard statement?Remember that the statement included thesentences, ‘Because Darwin's Theory is atheory, it continues to be tested as newevidence is discovered. The Theory is not afact.’ This implies that theories are somehowless certain than facts, or that theories are tobe given less credence simply because ‘theyare theories.’ It is on this basis thatevolutionary theory has been regarded bysome as open to sufficient doubt as to makethe inclusion of ID a viable scientificalterative to be considered in the classroom.But, as we have just seen, neither theories norhypotheses can be equated with facts. Thus,to say ‘this theory (or hypothesis) is just atheory, not a fact’ is a misnomer, for the claimdoes not properly recognize the fundamentalrelations between facts, and hypotheses andtheories.

Evolutionary biology has been profoundlysuccessful for the very reason that it servesthe goal of science – to provide avenues forthe ever-increasing acquisition ofunderstanding of facts. But, at any time wewish to critically assess the merits of anytheory or hypothesis, we always need to becareful that we use our words correctly.

THEORIES, SCIENTIFIC AND OTHERWISE:HOW DOES ONE DECIDE?

In the landmark court decision in 2005regarding the Dover PA school boardrequirement that ID be considered alongsideevolutionary biology in science classrooms,US District Judge John E. Jones III noted thatit is ‘readily apparent to the Court that ID failsto meet the essential ground rules that limitscience to testable, natural explanations.’While the court identified the seminalcriterion for demarcating science from non-science as activities whose goals are theacquisition of understanding, we often witnessa lack of clarity surrounding this separation.Indeed, this confusion exists not only in thelay community, but among some scientists aswell. Part of the confusion lies in the fact thatthe term theory is often only thought to residewithin the realm of science, such thatevolution is claimed to be a scientific theory,whereas ID is not a theory at all. The problemwith this reasoning is that it does not identifythat there can be non-scientific theories, ofwhich ID is one. Another part of theconfusion is that the tenets of demarcationbetween science and non-science are often notpresented by advocates of ID, or incorrectlypresented, and science teachers apparentlyhave not been particularly successful atstressing this demarcation.

We can readily identify the criterion thatseparates scientific from non-scientifictheories and hypotheses, giving us theopportunity to understand what is meant whenthe terms theory and hypothesis are applied,

-6-

and critically evaluated, in all fields ofscience, as opposed to fields such asmetaphysics and religion, to which IDbelongs.

Scientific and non-scientific theories - howare they different? Recall that we earlier defined a theory as ‘anestablished and accepted explanatory concept,or set of concepts, that we apply toobservations to give us understanding.’ Sucha definition allows for theories to function inboth scientific and non-scientific contexts, ashas been the case throughout history thathumans have applied a variety of theories toaccount for what they observe around them.

The criterion separating scientific andnon-scientific theories and hypotheses is thatof testing. The concept of testing traces backto the ancient Greeks, where one wishes tocritically determine whether or not someexplanatory paradigm successfullycharacterizes what occurs around us (or hasoccurred). If a theory or hypothesis ispresented in a form that makes it immune totesting, then it can not be regarded asappropriate for consideration in any field ofscience, since there can be no evidence,potential or otherwise, that could be sought torefute it. The mechanics of testing hypothesesand theories will be outlined next using twosimple examples. It is from these examplesthat we can identify ID as immune to testing,and as a result, lies outside the realm ofscience.

The mechanics of hypothesis testing.Let’s consider again our observationstatement, ‘Here is a glass of ice water’(Figure 1). Recall that the evidence fromwhich we might infer that there is a glass andice water on the table is that we applied to oursense perceptions at least two theories, glassand water. In other words, given our pastexperiences with use of these theories for

understanding our surroundings, we believethose to be the best available for giving usunderstanding of this object on the table.Recall, however, no theory or hypothesis isguaranteed to be true. While we constantlyattempt to explain, by the use of hypothesesand theories, what we perceive around us, noexplanation is fail safe. The process ofacquiring understanding is one of trial anderror. We try out theories and hypotheses andjudge to what extent they give usunderstanding. So how would one go aboutevaluating the ice water hypothesis?

If we assume the explanatory success ofour theories about glass and water, and thatour hypothesis is true, that there really is aglass of ice water before us, then there arespecific predictions we can derive from ourhypothesis. For instance, here are twopredictions that follow from the theories andhypothesis: the glass should feel very cold andthe ice and water should move freely in theglass. Notice that these conditions areexpected consequences if we are dealing withice water. We expect these consequencesbecause of what we know of the theory ofwater that we have applied to our observation.In essence, we have stated two possible testsof our hypothesis, and we are now in aposition to actually test it. You place yourhand around the glass and immediately noticeit is not cold. You move the glass and noticethat the water and ice don’t move. You eventurn the glass upside down, and nothing poursout (Figure 3). Our predictions are not met,indicating the hypothesis to be false. Theinitial explanation of our observations inFigure 1 is incorrect.

So where does this leave us? Well, youcontinue to assume you are not hallucinating,and you do not think your theories are in needof revision – they have well tested in the past.There is a glass on the table, but it is not filledwith ice water. Since you have shown yourhypothesis to be the wrong explanation, and

-7-

you now have additional informationregarding the substance in the glass (Fig. 3),you then entertain a new hypothesis: the glassis filled with lucite plastic made to look likeice water. In other words, at the point youdecided your original hypothesis is no longera worthy explanation, you applied anothertheory to your old and new observations, thetheory of lucite, from which you produced anew hypothesis.

Here we have the fundamental processused in all fields of science to test hypotheses.If a hypothesis is not open to being tested,meaning that specific predictions cannot beformulated and subsequent observationsmade, even potentially, then there is no meansto critically assess the utility of the hypothesisin our quest for understanding.

The mechanics of theory testing.The basic principles outlined above for thetesting of hypotheses also apply to the testingof theories in science. As a theory is a generalstatement regarding cause and effect relations,then the critical evaluation of a theory

requires determining the accuracy of thisclaimed relationship. Just as predictedconsequences serve as tests of hypotheses,such predictions are also the basis for testingtheories. Let’s consider a real example.

Albert Einstein’s theory of generalrelativity, published in 1915-16, states thatgravity is a manifestation of curved space andtime. The theory ushered in a new revolutionin physics and astronomy because it appearedto provide a broad explanatory framework,not only for what was already understood byway of Newtonian mechanics, but alsoaccounted for phenomena previously regardedas anomalies in light of those mechanics.

To test Einstein’s theory, that there is acausal relation between gravity and space-time, we would need to consider actualconditions where the actions of gravity couldbe observed. A fundamental prediction ofgeneral relativity is that rays of light, whichare weightless, should be deflected at aspecifiable angle in the presence of agravitational field. An opportunity to witnessjust such an event took place in 1919, with theoccurrence of a solar eclipse. At the point themoon completely blocks the sun, it would bepossible to determine the positions of starswho’s light passes close to the sun. Then,compare these positions with those when thesun is not in the field of view. If Einstein’stheory correctly describes the affect of gravityon space-time, then the deflection of light raysfrom the stars should be the angle predictedby the theory, as illustrated in Figure 4. Theresultant differences in eclipse and non-eclipse star positions was what was expected,thereby supporting the theory.

Theory and hypothesis testing: similar yetdifferent.What you might notice from the generalrelativity example is that testing a theory isessentially the act of performing anexperiment, whether occurring naturally, like

Figure 3. Our hypothesis, ‘Here is a glass of ice

water,’ has failed the tests!

-8-

the solar eclipse, or contrived by an individualin a laboratory. As theories state that undercertain initial or causal conditions one shouldexpect specific effects to follow if the theoryis true, the way to test a theory is to witnessthe initial conditions and observe whether ornot the resultant effects are as predicted. Thisis somewhat different from the testing ofhypotheses, which make claims that specificcausal events already occurred and what wehope to witness are predicted effects thatsupport a hypothesis. While the testing ofboth theories and hypotheses rely on thepredictions of consequences, the nature ofthose consequences do differ in content.

Recall from our hypothesis that there is aglass of ice water on the table (Fig. 1), weoffered a causal account that explains one’ssense data – that there is ice water in the glass.Since we were not present to witness the eventof filling the glass, or examine the substancewithin it, we have no direct evidence of thelink between cause and effect. So, to test thishypothesis we had to resort to predictingconsequences related as specifically aspossible to the suggested cause, but that areindependent of the observed effect. Thesituation in the testing of theories differs inthat we first observe, therefore know, theactual cause, and we then see what effectensues, and determine whether or not that

effect is what was predicted from the theory.

CONSEQUENCES FOR EVOLUTIONARY

BIOLOGY AND IDNow that we recognize the most fundamentalmechanics for critically evaluating scientifictheories and hypotheses, that of predictingcrucial, observable (or at least potentially so)consequences, we have the most importantbenchmark that distinguishes evolutionarybiology from ID theory. For any theory to beseriously entertained in the realm of science,one must at least be able to present thepotential tests from which the necessaryexperimental and/or causal regimes might beformulated. Are such requirements possiblefor evolutionary biology? For ID?

One of the misconceptions regarding whatis so often referred to as evolutionary ‘theory’is that there is not one grand, all-encompassing statement that can characterizesuch a ‘theory.’ Rather, evolution is not asingle theory, but a conglomeration ofinterconnected theories. This is readilyrecognized even in Charles Darwin’s 1859book, On the Origin of Species, where heshowed the interdependence of four classes ofevents needed to explain differences betweenorganisms over time: (i) organisms exhibitdiscernable variation; (ii) some of thisvariation is passed from parent(s) to offspring;

Figure 4. The deflection of starlight by the sun.

-9-

(iii) more offspring are produced than cansurvive and reproduce; (iv) those organismswith traits that enhance their survival andreproduction will leave offspring with thosetraits, i.e., there is a process of naturalselection. In the 20 century, with theth

development of fields such as genetics,population dynamics, and ecology, thefundamental principles comprisingevolutionary biology (not theory) can besummarized as follows [adapted from D.J.Futuyma (2005), Evolution]: (1) geneticvariation in the expressions of traits arises byrandom mutation and genetic recombination;(2) changes in the proportions of alleles(alternate forms of genes) and genotypes(genetic information) within a population mayresult in the presence of different genotypesbetween generations; (3) such changes in theproportions of genotypes may occur either byrandom fluctuations (‘genetic drift’) or by thenonrandom process of natural selection; and(4) due to the differential influences of geneticdrift and natural selection, the traits oforganisms in populations may diverge overtime.

We readily see that we are not dealingwith a single theory, but rather a set oftheories, each of which plays a vital part whenattempting to understand the vast past andpresent diversity of life. Especiallythroughout the 20 century, the principles ofth

genetic variation, population dynamics,ecology, natural selection, etc., have beensubjected to rigorous testing, and this processcontinues. To answer our question earlier inthis section – are the requirements for testingpossible for the various components ofevolutionary biology – the answer is ‘yes.’Whether laboratory experiments or fieldobservations, the theories scientists applyunder the heading of ‘evolution’ have beensubjected to critical scrutiny. Such testing hasbeen possible because science demands thatobservations be made of causes and checked

against the evidence provided by effects,predicted or otherwise. Just as the correctnessof any theory or hypothesis is neverguaranteed, scientists continually seekevidence to judge their evolutionary conceptsfor acquiring understanding.

Can we speak so affirmatively of ID? Cana theory that suggests some purposive force orentity has driven, is driving, and will drive thediversity of life on Earth be tested? Thisquestion goes back to the basis for this article:can an ID theory be subjected to the samecritical procedures of testing applied to alldisciplines of evolutionary biology, much lessall fields of science? As we saw earlier inspeaking of the testing of theories, the natureof such tests have the character ofexperiments, where one must be in a positionto witness both cause(s) and effects, such thatthe relations between the two can be judged aseither supporting or refuting a theory. Is itpossible to produce experimental conditions,much less natural conditions, that couldpotentially test any aspect of ID? Given thetenets of testing, it is clear that no area ofbiology could succeed at developing a validtest, much less produce test evidence thatcould support the theory as an alternative toevolution. For ID testing to be possible wewould have to witness initial, causalconditions so as to be able to empiricallyidentify the presence of the intelligent causeto which predicted effects could or could notbe associated. Unfortunately, there are noconceivable test conditions under which suchcause-effect relations can be discerned. What3

A more extensive discussion of the3

topic of testing a theory of intelligent designis provided in the companion article, TheMechanics of Testing a Theory: Implicationsfor Intelligent Design, available athttp://www.nhm.org/research/annelida/Mechanics_of_Testing.htm.

-10-

causal conditions one might witness wouldhave to be based on scientifically acceptablestandards, and the result would be that testconsequences would inevitably be explainedby natural, rather than supernaturalphenomena. ID does not lend itself to thekind of scrutiny required of scientific theories.As such, the beauty of ID is that it allows onethe luxury of explaining anything withimpunity. But, as the goal of science is toacquire ever increasing understanding throughcritical evaluation, ID is at odds with thatgoal, which immediately precludes it fromserious consideration in any realm of science.

CONCLUSIONS

What has been regarded as a debate over theteaching of ID as a viable scientific alternativeto evolutionary biology in public schoolclassrooms is really not a debate at all. Theissue at hand is not that of teaching twocompeting scientific theories, but rather anattempt to interject a loosely formulated,nonscientific theory into the realm ofscientific investigation. We have examinedtwo of the common misconceptions that havebeen used to sustain this action: the distinctionbetween fact and theory, and the requirementthat hypotheses and theories be open totesting. If the most fundamental hallmark ofall fields of science is that we want tocontinually subject our procedures forknowledge growth to the most criticalevaluative processes, then ID falls so far fromthe mark as to not be worthy of seriousconsideration. This by no means is a criticismof one’s desire to invoke ID. The criticismdeveloped in this article is that the ongoingattempts to associate ID with science, andscience education, are acts of misrepresentingthe nature of scientific inquiry. The purposeof a science classroom curriculum is to teachthe principles of doing science. The centralissue in the discussion of evolution versus IDis not whether or not ID should be considered.

The real issue is that all fields of science haveestablished the criterion of testing as thehallmark that separates scientific from non-scientific approaches to acquiring knowledge.While ID is a theory in its own right, it is nota scientific theory, and thus cannot beconsidered in the science classroom.

ACKNOWLEDGMENTS

I am grateful to Drs. Joel Martin and ÁngelValdés, Natural History Museum of LosAngeles County, for valuable comments.

About the author. J. Kirk Fitzhugh has beenCurator of Polychaetes at the Natural HistoryMuseum of Los Angeles County since 1990.He received his B.S. in marine biology fromTexas A&M University at Galveston, M.S. inbiology from Texas A&M University, andPh.D. in biology from George WashingtonUniversity. While his research has focusedmainly on the systematics of polychaeteworms around the world, Dr. Fitzhugh’sresearch interests extended to thephilosophical foundations of evolutionarybiology about 10 years ago.