Chapter 1

Models and Methods


 All that belongs to human understanding, in this deep ignorance and obscurity, is to be skeptical, or at least cautious; and not to admit of any hypothesis, whatsoever; much less, of any which is supported by no appearance of probability.

— David Hume

Lack of Evidence

Many theologians and theistic scientists claim that evidence has been found for the existence of the Judaic-Christian-Islamic God or, at least, some being with supernatural powers. However, they cannot deny that their evidence is not sufficiently convincing to satisfy the majority of scientists. Indeed, as we saw in the Preface, the overwhelming majority of prominent American scientists has concluded that God does not exist. If God exists, where is he? Philosopher Theodore Drange has termed this the lack-of-evidence argument, which he states formally as follows:


1.     Probably, if God were to exist, then there would be good objective evidence for his existence.

2.     But there is no good objective evidence for his existence.

3.     Therefore, probably God does not exist.


Drange criticizes premise 1 of the lack-of-evidence argument, pointing out that God could simply choose not to use the channel of objective evidence but directly implant that knowledge in human minds.[1] However, as he and others have pointed out, such a deity would not be a perfectly loving God and the very existence of nonbelievers in the world who have not resisted such belief is evidence against his existence.[2] The problem of divine hiddenness is one that has taxed the abilities of theologians over the years—almost as much as the problem of evil, which questions how an omnibenevolent, omnipotent, and omniscient God can allow so much unnecessary suffering among the planet’s humans and animals. We will return to each of these problems.

However, independent of the unknowable intentions of a hypothetical being of infinite power and wisdom, objective evidence for an entity with godlike attributes should be readily available. After all, God is supposed to play a decisive role in every happening in the world. Surely we should see some sign of that in objective observations made by eye and ear, and especially by our most sensitive scientific instruments.

The founders and leaders of major religions have always claimed that God can be seen in the world around us. In Romans 1:20, St. Paul says: “Ever since the creation of the world his invisible nature, namely his eternal power and deity, have been clearly perceived in the things that have been made.” We will look for evidence of God in the things that have been made.

The Nature of Scientific Evidence

Before examining specific data, let us consider what constitutes “scientific evidence.” Here I will limit myself to the kind of evidence that is needed to establish the validity of an extraordinary claim that goes beyond existing knowledge. Clearly the standard for this must be set much higher than that for an ordinary claim.

            For example, an ordinary claim might be that an 81-milligram aspirin taken daily will reduce the chance of heart attacks and strokes. Such a claim is ordinary, because we have a plausible mechanism for such an effect in the resulting slight thinning of the blood. By contrast, an extraordinary claim might be that such a therapy would cure AIDS. Lacking any plausible mechanism, we would have to demand far more confirmatory data than in the first case.

            We often hear of stories citing examples of dreams that came true. This would seem to suggest a power of the mind that goes beyond known physical capabilities. However, in this case, a strong selection process is taking place whereby all the millions of dreams that do not come true are simply ignored. Unless otherwise demonstrated, a plausible explanation that must first be ruled out is that the reported dream came true by chance selection out of many that had no such dramatic outcome.

            How can we rule out chance or other artifacts? This is what the scientific method is all about. We might do a controlled experiment with hundreds of subjects recording their dreams upon awaking every morning. Independent investigators, with no stake in the outcome one way or another, then perform a careful statistical analysis of the data. It would help if the dream outcomes were something simple and quantitative, like the winning number for a future lottery. Then the results could be compared with the easily calculated expectations from chance.

            Allow me to list a few of the rules that the scientific community conventionally applies when evaluating any extraordinary claim. This is not complete by any means; nowhere can we find a document that officially lays down the scientific method to the complete satisfaction of a consensus of scientists and philosophers. However, five conditions suffice for our evaluation of claims of empirical evidence for extraordinary empirical claims in science:

<b>Conditions for Considering Extraordinary Claims</b>


1.          The protocols of the study must be clear and impeccable so that all possibilities of error can be evaluated. The investigators, not the reviewers, carry the burden of identifying each possible source of error, explaining how it was minimized, and providing a quantitative estimate of the effect of each error. These errors can be systematic—attributable to biases in the experimental set up—or statistical—the result of chance fluctuations. No new effect can be claimed unless all the errors are small enough to make it highly unlikely that they are the source of the claimed effect.

2.          The hypotheses being tested must be established clearly and explicitly before data taking begins, and not changed midway through the process or after looking at the data. In particular, “data mining” in which hypotheses are later changed to agree with some interesting but unanticipated results showing up in the data is unacceptable. This may be likened to painting a bull’s-eye around wherever an arrow has struck. That is not to say that certain kinds of exploratory observations, in astronomy, for example, may not be examined for anomalous phenomena. But they are not used in hypothesis testing. They may lead to new hypotheses, but these hypotheses must then be independently tested according to the protocols I have outlined.

3.          The people performing the study, that is, those taking and analyzing the data, must do so without any prejudgment of how the results should come out. This is perhaps the most difficult condition to follow to the letter, since most investigators start out with the hope of making a remarkable discovery that will bring them fame and fortune. They are often naturally reluctant to accept the negative results that more typically characterize much of research. Investigators may then revert to data mining, continuing to look until they convince themselves they have found what they were looking for.[3] To enforce this condition and avoid such biases, certain techniques such as “blinding” may be included in the protocol, where neither the investigators nor the data takers and analyzers know what sample of data they are dealing with. For example, in doing a study on the efficacy of prayer, the investigators should not know who is being prayed for or who is doing the praying until all the data are in and ready to be analyzed.

4.          The hypothesis being tested must be one that contains the seeds of its own destruction. Those making the hypothesis have the burden of providing examples of possible experimental results that would falsify the hypothesis. They must demonstrate that such a falsification has not occurred. A hypothesis that cannot be falsified is a hypothesis that has no value.

5.          Even after passing the above criteria, reported results must be of such a nature that they can be independently replicated. Not until they are repeated under similar conditions by different (preferably skeptical) investigators will they be finally accepted into the ranks of scientific knowledge.

            Our procedure in the following chapters will be to select out, one by one, certain limited sets of attributes and examine the empirical consequences that can reasonably be expected by the hypothesis of a god having those attributes. We will then look for evidence of these empirical consequences.


Falsification was the demarcation criterion proposed in the 1930s by philosophers Karl Popper[4] and Rudolf Carnap[5] as a means for distinguishing legitimate scientific models from nonscientific conjectures. Since then, however, philosophers of science have found falsification insufficient for this purpose.[6] For example, astrology is falsifiable (indeed, falsified) and not accepted as science. Nevertheless, falsification remains a very powerful tool that is used whenever possible. When a hypothesis is falsifiable by a direct empirical test, and that test fails, then the hypothesis can be safely discarded.

Now, a certain asymmetry exists when testing scientific models. While failure to pass a required test is sufficient to falsify a model, the passing of the test is not sufficient to verify the model. This is because we have no way of knowing a priori that other, competing models might be found someday that lead to the same empirical consequences as the one tested.

Often in science, models that fail some empirical test are modified in ways that enable them to pass the test on a second or third try. While some philosophers have claimed this shows that falsification does not happen in practice, the modified model can be regarded as a new model and the old version was still falsified. I saw many proposed models falsified during my forty-year research career in elementary particle physics and astrophysics; it does happen in practice.[7]

Popper restricted falsification (which he equates to refutability) to empirical statements, and declared, “philosophical theories, or metaphysical theories, will be irrefutable by definition.”[8] He also noted that certain empirical statements are irrefutable. These are those statements that he called “strict or pure existential statements.” On the other hand “restricted” existential statements are refutable. He gives this example:


“There exists a pearl which is ten times larger than the next largest pearl.” If in this statement we restrict the words “There exists” to some finite region in space and time, then it may of course become a refutable statement. For example, the following statement is obviously empirically refutable: “At this moment and in this box here there exist at least two pearls one of which is ten times larger than the next largest pearl in this box.” But then this statement is no longer a strict or pure existential statement: rather it is a restricted existential statement. A strict or pure existential statement applies to the whole universe, and it is irrefutable simply because there can be no method by which it could be refuted. For even if we were able to search our entire universe, the strict or pure existential statement would not be refuted by our failure to discover the required pearl, seeing that it might always be hiding in a place where we are not looking.[9]


By this criterion, it would seem that the existence of God cannot be empirically refuted because to do so would require making an existential statement applying to the whole universe (plus whatever lies beyond). But, in looking at Popper’s example, we see this is not the case for God. True, we cannot refute the existence of a God who, like the pearl in Popper’s example, is somewhere outside the box, say, in another galaxy. But God is supposed to be everywhere, including inside every box. So when we search for God inside a single box, no matter how small, we should either find him, thus confirming his existence, or not find him thus refuting his existence.

Can Science Study the Supernatural?

Most national science societies and organizations promoting science have issued statements asserting that science is limited to the consideration of natural processes and phenomena. For example, the United States National Academy of Sciences has stated, “Science is a way of knowing about the natural world. It is limited to explaining the natural world through natural causes. Science can say nothing about the supernatural. Whether God exists or not is a question about which science is neutral.”[10]

Those scientists and science organizations that would limit science to the investigation of natural causes provide unwitting support for the assertion that science is dogmatically naturalistic. In a series of books in the 1990s, law professor Phillip Johnson argued that the doctrine that nature is “all there is” is the virtually unquestioned assumption that underlies not only natural science but intellectual work of all kinds.[11] In many of the public discussions we hear today, science is accused of dogmatically refusing to consider the possible role other than natural processes may play in the universe.

Given the public position of many scientists and their organizations, Johnson and his supporters have some basis for making a case that science is dogmatically materialistic. However, any type of dogmatism is the very antithesis of science. The history of science, from Copernicus and Galileo to the present, is replete with examples that belie the charge of dogmatism in science.

            What history shows is that science is very demanding and does not blindly accept any new idea that someone can come up with. New claims must be thoroughly supported by the data, especially when they may conflict with well-established knowledge. Any research scientist will tell you how very difficult it is to discover new knowledge, convince your colleagues that it is correct—as they enthusiastically play devil’s advocate—and then get your results through the peer-review process to publication. When scientists express their objections to claims such as evidence for intelligent design in the universe, they are not being dogmatic. They are simply applying the same standard they would for any other extraordinary claim and demanding extraordinary evidence.

            Besides, why would any scientist object to the notion of intelligent design or other supernatural phenomena, should the data warrant that they deserve attention? Most scientists would be delighted at the opening up of an exciting new field of study that would undoubtedly receive generous funding. As we will see, intelligent design, in its current form, simply incorporates neither the evidence nor the theoretical arguments to warrant such attention.

Furthermore, the assertions that science does not study the supernatural and that supernatural hypotheses are untestable are factually incorrect. Right under the noses of the leaders of national science organizations who make these public statements, capable, credentialed scientists are investigating the possibility of supernatural causes. As we will discuss in a later chapter, reputable institutions such as the Mayo Clinic, Harvard University, and Duke University are studying phenomena that, if verified, would provide strong empirical support for the existence of some nonmaterial element in the universe. These experiments are designed to test the healing power of distant, blinded intercessory prayer. Their results have been published in peer-reviewed medical journals.

Unfortunately, the prayer literature is marred by some very poor experimental work. But reading the best of the published papers of the most reputable organizations you will witness all the indications of proper scientific methodology at work. If they are not science, then I do not know what is.

The self-imposed convention of science that limits inquiry to objective observations of the world and generally seeks natural accounts for all phenomena is called methodological naturalism. We have also noted that methodological naturalism is often conflated with metaphysical naturalism, which assumes that reality itself is purely natural, that is, composed solely of material objects.

Methodological naturalism can still be applied without implying any dogmatic attachment to metaphysical naturalism. The thesis of this book is that the supernatural hypothesis of God is testable, verifiable, and falsifiable by the established methods of science. We can imagine all sorts of phenomena that, if observed by means of methodological naturalism, would suggest the possibility of some reality that is highly unlikely to be consistent with metaphysical naturalism.

For example, it could happen that a series of carefully controlled experiments provide independent, replicable, statistically significant evidence that distant, intercessory prayer of a specific kind, say Catholic, cures certain illnesses while the prayers of other religious groups do not. It is difficult to imagine any plausible natural explanation for this hypothetical result.

Impossible Gods

Before proceeding with the scientific evidence bearing on the God hypothesis, let us make a quick review of those disproofs of God’s existence that are based on philosophy. For a recent survey, see The Non-existence of God by Nicholas Everitt.[12] Philosophers Michael Martin and Ricki Monnier have assembled a volume of essays on the logical arguments claiming to show the impossibility of gods with various attributes.[13] Here is how they classify these types of disproofs:


q  definitional disproofs based on an inconsistency in the definition of God;

q  deductive evil disproofs based on the inconsistency between the existence of God who has certain attributes and the existence of evil;

q  doctrinal disproofs based on an inconsistency between the attributes of God and a particular religious doctrine, story, or teaching about God;

q  multiple-attribute disproofs based on an inconsistency between two or more divine attributes;

q  single-attribute disproofs based on an inconsistency within just one attribute


These disproofs merit greater credence than the claimed philosophical proofs of the existence of God, for the same reason scientists and philosophers give more credence to falsifications of scientific models than to the verifications. The logical disproofs seem inescapable, unless you change the rules of the game or, more commonly, change the definitions of the words being used in the argument.

In the following, formal statements for a sample of nonexistence arguments are listed, just to give the reader the flavor of the philosophical debate. They will not be discussed here since they are independent of the scientific arguments that form my main thesis; the conclusions of this book are in no way dependent on their validity. They are listed for completeness and for contrast with the scientific arguments. For the details, see the individual essays in the compilation by Martin and Monnier.[14]

The first two are examples of definitional disproofs:

<b>An All-Virtuous Being Cannot Exist</b>


1.     God is (by definition) a being than which no greater being can be thought.

2.     Greatness includes the greatness of virtue.

3.     Therefore, God is a being than which no being could be more virtuous.

4.     But virtue involves overcoming pains and danger.

5.     Indeed, a being can only be properly said to be virtuous if it can suffer pain or be destroyed.

6.     A God that can suffer pain or is destructible is not one than which no greater being can be thought.

7.     For you can think of a greater being, one that is nonsuffering and indestructible.

8.     Therefore, God does not exist.[15]


<b>Worship and Moral Agency</b>


1.     If any being is God, he must be a fitting object of worship.

2.     No being could possibly be a fitting object of worship, since worship requires the abandonment of one’s role as an autonomous moral agent.

3.     Therefore, there cannot be any being who is God.[16]


We have already briefly noted the problem of evil, and will be saying much more about it. For now, let us just indicate its formal statement:

<b>The Problem of Evil</b>


1.     If God exists, then the attributes of God are consistent with the existence of evil.

2.     The attributes of God are not consistent with the existence of evil.

3.     Therefore, God does not and cannot exist.[17]


The following three are examples of multiple-attribute disproofs:

<b>A Perfect Creator Cannot Exist</b>


1.     If God exists, then he is perfect.

2.     If God exists, then he is the creator of the universe.

3.     If a being is perfect, then whatever he creates must be perfect.

4.     But the universe is not perfect.

5.     Therefore, it is impossible for a perfect being to be the creator of the universe.

6.     Hence, it is impossible for God to exist.[18]


<b>A Transcendent Being Cannot Be Omnipresent</b>


1.     If God exists, then he is transcendent (i.e., outside space and time).

2.     If God exists, he is omnipresent.

3.     To be transcendent, a being cannot exist anywhere in space.

4.     To be omnipresent, a being must exist everywhere in space.

5.     Hence it is impossible for a transcendent being to be omnipresent.

6.     Therefore it is impossible for God to exist.[19]


<b>A Personal Being Cannot Be Nonphysical</b>


1.     If God exists, then he is nonphysical.

2.     If God exists, then he is a person (or a personal being).

3.     A person (or personal being) needs to be physical.

4.     Hence, it is impossible for God to exist.[20]


Finally, here is an example of a single-attribute disproof:

<b>The Paradox of Omnipotence</b>


1.     Either God can create a stone that he cannot lift, or he cannot create a stone that he cannot lift.

2.     If God can create a stone that he cannot lift, then he is not omnipotent.

3.     If God cannot create a stone that he cannot lift, then he is not omnipotent.

4.     Therefore God is not omnipotent.[21]


The reader will undoubtedly see much in these bare formal statements that needs clarification; again I address you to the original essays for details and additional disproofs of this kind. Like most philosophical discussions, it mainly comes down to the meanings of words and assembling them into coherent, consistent statements. The philosophers who formulated these disproofs have been careful about defining the terms used, while those who dispute them will generally disagree with those definitions or the way they have been interpreted. As a result, the debate continues.

Ways Out

Ways out of purely logical arguments can always be found, simply by relaxing one or more of the premises or, as noted, one of the definitions. For example, assume God is not omnibenevolent. Indeed, the God of the more conservative elements of Judaism, Christianity, and Islam that take their scriptures literally can hardly be called omnibenevolent—or even very benevolent. No one reading the Bible or Qur’an literally can possibly regard the God described therein as all-good. We will see examples later, but for now the reader is invited to simply pick up an Old Testament or Qu’ran, open to a random page, and read for a while. It will not take you long to find an act or statement of God that you find inconsistent with your own concepts of what is good. And, as we will also see, much in the Gospel can hardly be called “good.”

In any event, the scientific case is not limited to an omnibenevolent, omniscient, or omnipotent god.

The scientific method incorporates a means to adjudicate disputes that otherwise might run in circles, never converging as disputants on all sides of an issue continually redefine and refine their language. In science we are able to break out of this vicious cycle by calling upon empirical observations as the final judge. Of course, ways out of the scientific arguments can also be achieved by redefining God or by disputing the empirical facts. The reader will simply have to judge for herself whether the examples I present are convincing.

Models and Theories

Science is not just a matter of making observations but also developing models to describe those observations. In fact, philosophers have pointed out that any observation or measurement we make in science depends on some model or theory. They assert that all observations are “theory-laden.” For example, when we measure the time it takes for a particle to move from one point to another we must first assume a model in which particles are visualized as moving in space and time. The model must begin by defining space and time.

The use of models, which are simplified pictures of observations, is not limited to the professional practice of science. They are often used to deal with the ordinary problems of life. For example, we model the sun as an orb rising in the east and setting in the west. Travelers heading to the west can point themselves each day in the direction of the setting sun and, correcting for some northward or southward drift (depending on season), arrive safely at their destination. No additional elements to the model are needed—in particular, no metaphysics. The ancient Greeks viewed the sun as the gold-helmeted god Apollo, driving a golden chariot across the sky. The ancient Chinese thought it was a golden bird. Neither metaphysical model offers any additional aid to our travelers in their navigation. And, that lack of necessity in the absence of any other evidence testifies strongly for the nonexistence of such a god or golden bird.

While utilizing models is a normal process in everyday life, scientific models objectify and, whenever possible, quantify the procedure—thus providing a rational means for distinguishing between what works and what does not. Whenever possible, mathematics and logic are used as tools to enforce a consistency that is not always found in commonplace statements, which are formulated in the vernacular. For example, instead of saying that your blood pressure is probably high, a physician will measure it and give you two numbers, say 130 over 100. Then he might prescribe some calculated amount of medication to bring the 100 down to 80.

Scientific instruments that enhance the power of our senses commonly yield quantifiable measurements, enabling scientists to deal with variables having numerical values upon which all observers can agree—within equally quantifiable measurement errors. While some sciences may deal with nonnumerical variables, physical models are almost always quantitative and the logical power of mathematics put to great use in their utilization.

Most scientific models begin by defining their observables operationally, that is, by characterizing them in terms of a well-prescribed, repeatable measuring procedure. For example (as Einstein emphasized), time is defined as what you read on a clock. Temperature is what you read on a thermometer. Specific instruments are chosen as standards. A mathematical framework is then formulated that defines other variables as functions of the observables and postulates connections between these quantities.

The term model usually applies to the preliminary stages of a scientific process when considerable testing and further work still need to be done. The “theories” that arise from this effort are not the unsupported speculations that they are often accused of being by those unfamiliar with the scientific method or by those wishing to demean it. To be accepted into the ranks of scientific knowledge, theories must demonstrate their value by passing numerous, risky empirical tests and by showing themselves to be useful. Theories that fail these tests, or do not prove useful, are discarded.

In this book we will make frequent reference to the standard models of fundamental physics and cosmology. By now these have sufficiently advanced to the level where they can be honestly recognized as standard theories, although their prior designations as models continue to be used in the literature, presumably to maintain familiarity. I find it amusing and ironic that opponents of evolution think they are undermining it by calling it “just a theory.”

The validity of the scientific method is justified by its immense success. However, we must recognize and acknowledge that scientific models and theories, no matter how well established, are still human contrivances and subject to change by future developments. This is in contrast to revelations from God, which should be true unconditionally and not subject to revision. Furthermore, the elements of scientific models, especially at the deepest level of quantum phenomena, need not correspond precisely to the elements of whatever “true reality” is out there beyond the signals we receive with our senses and instruments. We can never know when some new model will come along that surpasses the old one. We regard such a happening as the welcome progress of science rather than some disastrous revolution that tears down the whole prior edifice, rendering it worthless. For example, despite a common misunderstanding, the models of Newtonian mechanics were hardly rendered useless by the twin twentieth-century developments of relativity and quantum mechanics. Newtonian physics continue to find major application in contemporary science and technology. It is still what most students learn in physics classes and what most engineers and others use when they apply physics in their professions.

Perhaps quarks and electrons are not real, although they are part of the highly successful standard model of particle physics. We cannot say. But we can say, with high likelihood, that some of the elements of older models, such as the ether, are not part of the real world. And, while we cannot prove that every variety of god or spirit does not exist in a world beyond the senses, we have no more rational basis for including them than we have for assuming that the sun is a god driving a chariot across the sky. Furthermore, we can proceed to put our models to practical use without ever settling any metaphysical questions. Metaphysics has surprisingly little use and would not even be worth discussing if we did not have this great desire to understand ultimate reality as best as we can.

The ingredients of scientific models are not limited to those supported by direct observation. For example, the standard model of elementary particles and forces contains objects such as quarks, the presumed constituents of atomic nuclei, which have never been seen as free particles. In fact, the theory in its current form requires that they not be free. The observation of a free quark would falsify that aspect of the standard model, although nicely confirm the quark idea itself.

Indeed, the development of models in physics is often motivated by considerations of logical and mathematical beauty, such as symmetry principles. But they still must be tested against observations.

            Astronomical models include black holes, which can only be observed indirectly. Cosmological models include dark matter and dark energy, which remain unidentified at this writing but are inferred from the data. The models currently used in modern physics, astronomy, and cosmology are solidly grounded on direct observations and have survived the most intensive empirical testing. By virtue of this success, they can be used to make inferences that are surely superior to speculations simply pulled out of thin air.

            Physicists generally speak as if the unobserved elements of their models, such as quarks, are “real” particles. However, this is a metaphysical assumption that they have no way of verifying and, indeed, have no real need (or desire) to do so. The models of physics and their unobserved elements are human inventions and represent the best we can do in describing objective reality. When a model successfully describes a wide range of observations, we can be confident that the elements of those models have something to do with whatever reality is out there, but less confident that they constitute reality itself.

            On the other hand, if a model does not work there is no basis to conclude that any unique element of that model is still part of reality. An example is the electromagnetic ether, which was discussed earlier.

            Having read this, please do not assume that the doctrine of postmodernism is being promoted here. Science is decidedly not just another cultural narrative. The science referred to is called “Western science,” which was developed originally by Europeans utilizing mathematical insights from India (the concept of “zero”), the Arab world (numerals, algebra), and other cultures. Peoples in all but the most primitive societies now utilize science. While we might consider science another “cultural narrative,” it differs from other cultural narratives because of its superior power, utility, and universality.

Modeling God

Everyone involved in discourses on the existence of gods may be well advised to consider the approach outlined above. Like quarks, the gods are human inventions based on human concepts. Whether or not we can say if the gods people talk about have anything to do with whatever objective reality is out there depends on the empirical success of the models that are built around these hypothetical entities. Whatever a god’s true nature, if one exists, a god model remains the best we can do in talking about that god.

            If we accept this procedure, then we can eliminate a whole class of objections that are made to types of logical and scientific arguments formulated in this book. In these arguments, God is assumed to have certain attributes. The theologian may ask: how can we mere mortals know about the true nature of a god who lies beyond our sensibilities? The answer is that we do not need to know—just as physicists do not need to know the ultimate reality behind quarks. Physicists are satisfied that they have a model, which currently includes quarks, that agrees beautifully with the data. The quark model is empirically grounded. It represents the best we humans have been able to do thus far in describing whatever objective reality underlies nuclear and subnuclear observations. Whether quarks are real or not does not change this. Whether any of the objects of scientific models are real or not does not change the fact that those models have immense utility. This includes Newtonian physics, despite the further developments of twentieth-century relativity and quantum mechanics.

Analogously, if a particular god model successfully predicts empirical results that cannot be accounted for by any other known means, then we would be rational in tentatively concluding that the model describes some aspect of an objective reality without being forced to prove that god really is as described in the details of the model.

Still, any god model remains a human invention, formulated in terms of human qualities that we can comprehend, such as love and goodness. Indeed, the gods of ancient mythology—including the Judaic-Christian-Islamic God—are clearly models contrived by humans in terms people could understand. What is amazing is that in this sophisticated modern age so many still cling to primitive, archaic images from the childhood of humanity.

            On the flip side, when a model is strongly falsified by the data, then those elements of the model that have been severely tested by observations should be rejected as not very likely to be representative of an objective reality.

The following example should illustrate this rather subtle concept. Observations of electromagnetic phenomena support a model of electromagnetism containing pointlike electric charges we can call electric monopoles. Examples include ions, atomic nuclei, electrons, and quarks. Symmetry arguments would lead you to include in the model point magnetic charges—magnetic monopoles.

Yet the simplest observed magnetic sources are described as magnetic dipolesbar magnets that have north and south poles. Electric dipoles such as hydrogen atoms, with a positive and negative point charge separated in space, exist as well. But you can tear them apart into separate electric monopoles, such as an electron and proton. On the other hand, if you cut away a piece of the north pole of a bar magnet, instead of getting a separate north and south monopole you get two dipoles—two bar magnets.

            Despite these empirical facts, some theoretical basis exists for magnetic monopoles, and they have been searched for extensively with no success. The current standard model contains perhaps a single magnetic monopole in the visible universe, which has no effect on anything. That is, the model does include a magnetic monopole, but we can proceed to use our conventional electromagnetic theory, which contains no magnetic monopoles, for all practical applications.

Let us apply this same line of reasoning to God. When we show that a particular model of God fails to agree with the data, then people would not be very rational in using such a model as a guide to their religious and personal activities. While it remains possible that a god exists analogous to the lonely magnetic monopole, one who has no effect on anything, there is no point worshiping him. The gods we will consider are important elements of scientific models that can be empirically tested, such as by the successful consequences of prayer.

The Scientific God Model

So, let us now define a scientific God model, a theory of God. A supreme being is hypothesized to exist having the following attributes:


1.     God is the creator and preserver of the universe.

2.     God is the architect of the structure of universe and the author of the laws of nature.

3.     God steps in whenever he wishes to change the course of events, which may include violating his own laws as, for example, in response to human entreaties.

4.     God is the creator and preserver of life and humanity, where human beings are special in relation to other life-forms.

5.     God has endowed humans with immaterial, eternal souls that exist independent of their bodies and carry the essence of a person’s character and selfhood.

6.     God is the source of morality and other human values such as freedom, justice, and democracy.

7.     God has revealed truths in scriptures and by communicating directly to select individuals throughout history.

8.     God does not deliberately hide from any human being who is open to finding evidence for his presence.


Most of these attributes are traditionally associated with the Judaic-Christian-Islamic God, and many are shared by the gods of diverse religions. Note, however, that the traditional attributes of omniscience, omnipotence, and omnibenevolence—the 3O characteristics usually associated with the Judaic-Christian-Islamic God—have been omitted. Such a God is already ruled out by the arguments of logical inconsistency summarized above. While the 3Os will show up on occasion as supplementary attributes, they will rarely be needed. For example, the case against a creator god will apply to any such god, even an evil or imperfect one. Furthermore, as will be emphasized throughout, the God of the monotheistic scriptures—Old Testament or Hebrew Bible, New Testament, and Qur’an—is not omnibenevolent, and so not ruled out by logical inconsistency. The observable effects that such a God may be expected to have are still testable by the normal, objective processes of science.

The Generic Argument

The scientific argument against the existence of God will be a modified form of the lack-of-evidence argument:


1.     Hypothesize a God who plays an important role in the universe.

2.     Assume that God has specific attributes that should provide objective evidence for his existence.

3.     Look for such evidence with an open mind.

4.     If such evidence is found, conclude that God may exist.

5.     If such objective evidence is not found, conclude beyond a reasonable doubt that a God with these properties does not exist.


Recall that it is easier to falsify a hypothesis than verify one. The best we can do if the data support a particular god model is acknowledge that faith in such a God is rational. However, just as we should not use a failed physical model that does not work, it would be unwise for us to guide our lives by religions that worship any gods that fail to agree with the data.



[1]Theodore M. Drange, Nonbelief and Evil: Two Arguments for the Nonexistence of God (Amherst, NY: Prometheus Books, 1998), p. 41.

[2] See also John L. Schellenberg, Divine Hiddenness and Human Reason (Ithaca, NY: Cornell University Press, 1993).

[3] For a good example of data mining, see my discussion of the experiment by Elisabeth Targ and collaborators in Victor J. Stenger, Has Science Found God? The Latest Results in the Search for Purpose in the Universe (Amherst, NY: Prometheus Books, 2003), pp. 250-53.

[4] Karl Popper, The Logic of Scientific Discovery, English edition (London: Hutchinson; New York: Basic Books, 1959). Originally published in German (Vienna: Springer Verlag, 1934).

[5] Rudolf  Carnap, “Testability and Meaning,” Philosophy of Science B 3 (1936): 19-21, B 4 (1937): 1-40.

[6] Philip J. Kitcher, Abusing Science: The Case Against Creationism (Cambridge, MA: MIT Press, 1982). Note that the author was refuting the common creationist claim that evolution is not science because it is not falsifiable. Kitcher need not have bothered. Evolution is eminently falsifiable, as we show in chapter 3.

[7] I discuss several examples in Victor J. Stenger, Physics and Psychics: The Search for a World Beyond the Senses (Amherst, NY: Prometheus Books, 1990).

[8] Karl Popper, “Metaphysics and Criticizability” in: Popper Selections, David Miller, ed. (Princeton, NJ: Princeton University Press, 1985). Originally published in 1958, p. 214.

[9] Ibid.

[10] National Academy of Sciences. Teaching About Evolution and the Nature of Science. Washington, DC: National Academy of Sciences, 1998. Online athtto:// . (Accessed March 5, 2006), p. 58.

[11] Phillip E. Johnson, Evolution as Dogma: The Establishment of Naturalism (Dallas, TX: Haughton Publishing Co., 1990); Darwin on Trial (Downers Grove, IL: InterVarsity Press, 1991); Reason in the Balance: The Case Against Naturalism in Science, Law, and Education (Downers Grove, IL: InterVarsity Press, 1995); Defeating Darwinism by Opening Minds (Downers Grove, lL: InterVarsity Press, 1997); The Wedge of Truth: Splitting the Foundations of Naturalism (Downers Grove, IL: InterVarsity Press, 2001).

[12] Nicholas Everitt, The Non-Existence of God (London, New York: Routledge, 2004).

[13] Michael Martin and Ricki Monnier, eds., The Impossibility of God (Amherst, NY: Prometheus Books, 2003).

[14] Ibid.

[15] Douglas Walton, “Can an Ancient Argument of Carneades on Cardinal Virtues and Divine Attributes Be Used to Disprove the Existence of God?, Philo 2, no. 2 (1999): 5-13; reprinted in Martin and Monnier 2003, pp. 35-44.

[16] James Rachels, “God and Moral Autonomy,” from James Rachels, Can Ethics Provide Answers? And Other Essays in Moral Philosophy (New York: Rowman & Littlefield Publishers, 1997), pp. 109-23; reprinted in Martin and Monnier 2003, pp. 45-58.

[17] Martin and Monnier 2003, p. 59.

[18] Theodore M. Drange, “Incompatible-Properties Arguments—A Survey, Philo 1, no. 2 (1998): 49-60; “ in Martin and Monnier 2003, pp. 185-97.

[19] Ibid.

[20] Ibid.

[21] J.L. Cowen, “The Paradox of Omnipotence Revisited,” Canadian Journal of Philosophy  3, no. 3 (March 1974): 435-45; reprinted in Martin and Monnier 2003, p. 337.