What is science? Is the use of animal models for the study of human disease and treatment an application of science? This essay will explore these questions.
Americans, Europeans, and Japanese For Medical Advancement assert:
- The results from experiments on animals are not predictive of what will occur in humans.
- By misleading scientists, the extrapolation of results from animal models harms human patients, indirectly by delaying life-saving discoveries, and directly by endangering human lives.
- Discoveries made from past experiments on animals could have occurred without animals. Animals are subject to the same laws of physics, chemistry and biology as humans, but were not needed to prove such concepts.
- Experiments on animals waste time, money, and personnel.
We base our assertions on:
- An analysis of what science is and an examination of the philosophy of science upon which biomedical research is based.
- A historical analysis of the process of medical discovery.
- An examination of current research projects.
- Multiple examples of the results of animal models.
This essay will focus on number 1. Books such as Sacred Cows and Golden Geese, The Human Harm from Animal Experiments and brochures available from AFMA provide support for numbers 2-4.
We will prove in this essay that the animal model paradigm is not based on science and should be abolished.
The Philosophy of Science Intrinsic to the very philosophy of modern biological science is the prediction that animal models of human disease will not be tenable. We will explore this facet of modern biology as it relates to our claims.
We will examine the epistemology, methodology and science of using animals as models of human disease. This examination will reveal that using animal models to study human diseases violates the philosophical underpinnings of science itself. As such, the study of animal models, though represented as science, is in fact pseudoscience, because the paradigm of animal models violates the criteria that form the foundation for true science.
We will not discuss what actually defines science in the detail one might find in a graduate course in the philosophy of science. (We will not examine for example, Wittgenstein's duck-rabbit problem here or other issues of the philosophy of science. Science is certainly theory-laden but for the purpose of this discussion we need not address issues such as that. For anyone wishing further study of the philosophy of science, we recommend Curd and Cover's Philosophy of Science, published by Norton in 1998.) We will engage in a basic review of the philosophy of science, however and we will simply define science for the reader as most scientists define it. We will leave out much that is controversial about the philosophy of science itself. We note, for the record, that science is not the only way to know reality. There are other ways of knowing reality: intuition, religion or spirituality, and aesthetics. However, in the estimation of both vivisectors (those who experiment upon animals) and AFMA, science is the chief means of knowing the reality of the material world.
All agree that the power of science lies in its ability to relate cause and effect. According to E. O. Wilson writing in his book, Consilience, “Science is the organized, systematic enterprise that gathers knowledge about the world and condenses the knowledge into testable laws and principles.” In practice, science is the observation, identification, description, experimental investigation, and theoretical explanation of natural phenomena.” It demands systematic methodology and study.
Critical Discourse vs. Experimentation
Before concepts reach a test phase, opinions and counter-opinions are offered in dispute. This is called critical discourse. Historically, before we could test certain concepts or phenomenon critical discourse was the final explanation for the concept or phenomenon. It is still the final word in arguments when we cannot test theories (Does God exist?) or when we do not have the data to form an intelligent hypothesis. (Why do women suffer from connective tissue diseases more than men?) The philosophy of science states “all disagreements about matters of fact are, in principle, open to rational clarification and resolution.”[1]
Science can move understanding from opinion/critical discourse – albeit opinion based on observation, logic and rational thought – to fact, whether partial or comprehensive. Kuhn called the shift of opinion to fact through science a move from critical discourse to experimentation. In his view, critical discourse occurs only during times of crisis, when the basis from which experimentation might proceed appears flawed.
By Kuhn's definition, science distinguished itself in its ability to solve puzzles. (How does HIV infect the cell? What chemical can we use to cure infections?) He suggested that in order for a field to be called science, its “conclusions must be logically derivable from shared premises.”[2]
Philosophers such as Popper, Kuhn, Lakatos and others have written about conditions that distinguish science from pseudoscience. Many criteria can be used, some more successfully than others. For an enterprise to be considered science, these conditions or demarcation criteria must be met. Demarcation criteria may include predictability, consilience, economy and other factors that we will explore in coming paragraphs. If a discipline fails to meet these criteria, then it is not science. It may be useful for other reasons and it may lead to truth, but it is not science. Whereas pseudoscience may occasionally achieve results, that does not overcome its inability to meet the demarcation criteria.
Falsifiability, Testability and Refutability
Karl Popper distinguished science from pseudoscience not owing to methodology but rather falsifiability. Einstein’s theory of relativity predicted that light would bend during an eclipse. His theory was tested and proven by results that could have been different from that which was predicted. The results could have been different and thus the theory could have been proven wrong or falsified. This stands in contradistinction to testing your horoscope’s prediction of what kind of day you are going to have. That prediction is so vague that many events of the day could be interpreted as having fulfilled the prediction. In other words, very strict criteria must be met when offering a proof. There must be risk involved when offering a proof. The theory cannot be so general as to be encompassed regardless of the results of the experiment. Popper thought:
- It is easy to find data to support a theory.
- Confirmation of the theory by prediction should count only if the prediction was risky. The prediction, and then proof of the theory by actually doing experiments to test the prediction does in part, at least, confirm the theory.
- Good theories not only predict happenings but also prohibit things from happening.
- A non-refutable theory is not scientific.
- Confirmatory evidence should not count as such unless it came about as an actual test, implying that the result could have been different.
- If a theory can be saved only by adding further assumptions it may still be true but “Occam's razor” applies. Occam's razor is the premise that if 2 theories explain a phenomenon equally well, then the one with the fewer assumptions or anomalies is true.
Thus, science differs from pseudoscience in many ways, including a theory’s ability to be proven false. Testability and refutability also refer to falsifiability. Astrology and other pseudoscientific endeavors escape falsification only by sacrificing falsifiability. In other words, they escape being proven false only by avoiding testing, or by being tested but making sure the test involves no risk, or when they are tested and proven nonscientific, by simply denying that the results falsify their theory.
Regardless of how one defines science, one usually gets back to falsifiability. Religious beliefs such as reincarnation, virgin birth, miracles, etc. are not falsifiable. That is not to say they are not true, only that believing in them is not science-based. Science is falsifiable. It has been tested millions of times in the form of technology. Science predicts that a machine can be made that will test how much iron a person has in her blood and sure enough, the machine is built and gives reliable, predictable data. Science is the best philosophy for understanding the material world because it has had the best track record when tested.
According to Kuhn a good scientific theory fulfills five criteria:
- It is accurate.
- It is both internally consistent and consistent with other knowledge of the time.
- It should have a broad scope. That is it should have implications for things beyond that which it was originally designed to explain.
- It should be simple.
- It should be fruitful, that is, it should yield new knowledge.[3]
Predictability vs. Verifiability
Both science and pseudoscience explain after the fact. Theories in each are verifiable. However, only science predicts. Science explains after the fact also but actively predicts a certain outcome that a test will either prove or disprove.[4] Some would call this prospective explanation versus retrospective explanation.
Theory, Practitioners, and Historical Context
Science looks for regularities and calls them laws or theories. The universe does seem to behave regularly in some cases, e.g. gravity. These laws are then offered to explain natural phenomena and to predict future natural phenomena. It is important to emphasize that science is tentative not dogmatic. It only offers the best solution for the problem, as best the current data can provide. A better or more comprehensive explanation of reality may come along. Later discoveries often change the paradigm, as happened with physics in the early twentieth century.
Paul Thagard also differentiated science from pseudoscience using three factors:
- Theory - structure, prediction, explanation, problem solving, physical foundation, etc.
- The community of advocates of the theory. Are the practitioners in agreement on the principles of the theory and how to go about solving the problems that the theory faces? Are they trying to explain the anomalies and do they consider the anomalies important? Are they comparing the success for their theory to the success of competing theories? Is the community actively trying to prove/disprove their theory?
- Historical context. A theory is rejected when a better one replaces it, or it has failed over a long period of time to explain the anomalies and hence is considered useless. Thus we must evaluate whether or not the theory has explained new facts and dealt with anomalies and how it stacks up against alternative theories.
Thagard proposes that a putative scientific theory be deemed pseudoscientific "if and only if:
- It has been less progressive than alternative theories over a long period of time and faces many unsolved problems. But,
- The community of practitioners makes little attempt to develop the theory towards solutions of the problems, shows no concern for attempts to evaluate the theory in relation to others, and is selective in considering confirmations and disconfirmations." [5]
He states, "Progressiveness is a matter of the success of the theory in adding to its set of facts explained and problems solved." He adds that pseudoscience relies on resemblance for much of its basis instead of causal relationships. (e.g., rhino horn for treating impotence. Mars is red like blood, so those born under the sign of Mars are more war-like than others.) Pseudoscience is also riddled by complex and ad hoc hypotheses thus violating the rule of parsimony or Occam's razor. Thagard's realization that not all activities that claim to be science are, is important because, as he states, "…society faces the twin problems of lack of public concern with the important advancement of science, and the lack of public concern with the important ethical issues now arising in science and technology, for example around the topic of genetic engineering. One reason for the dual lack of concern is the wide popularity of pseudoscience and the occult among the general public. Elucidation of how science differs from pseudoscience is the philosophical side of an attempt to overcome public neglect of genuine science." (Emphasis added)[6] We believe this explanation also applies to the question we are examining here: is the use of animal models for studying human disease good science?
We would add that another reason the public is dissuaded from scientific arguments is the fact that science is more difficult to comprehend than many issues that confront us in our everyday lives. Culturally, we are geared toward escapism and entertainment, not critical thinking. Our educational system has not emphasized critical thinking, and, hence, people do not understand the importance of science and are not prepared to think through scientific arguments. Further, the move toward specialization in science fields keeps decision making remote.
Paradigms and Science
Lakatos defined science as a paradigm with 1) a hard core of beliefs surrounded by 2) circles of less tenaciously held beliefs. Numbers 1 and 2 are encircled by 3) a heuristic ring. The core stays the same while the outer rings change. One experiment cannot prove or disprove a scientific paradigm; rather new data is always interpreted in light of the core. Paradigm shifts occur as when Einstein and modern physics supplanted Newtonian physics. But even then relativity did not do away with the laws of motion; it only modified them. The paradigm is subject to modification as new knowledge is added. The outer circles are constantly in flux with additions, negations and modifications.
According to Lakatos, all successful paradigms predict novel facts that could not have been predicted without them. He adamantly rejects the view that truth is whatever the majority believes, just because the majority believes it, e.g. the earth is flat, geocentricism - the earth is the center of the universe, or animal-models are useful for curing human disease because the vested interest groups so state.[7]
Respectability
- The study of and verification by scientific theory is most respected. This includes methodology and epistemology. Much of what Einstein predicted he did so based on theory without experimentation. Sir Arthur Eddington stated, "It is also a good rule not to put too much confidence in observational results until they are confirmed by theory." This essay concerns scientific theory.
- Examples from observation and controlled experiments that refute or support the theory are second.
- Statements made by experts are a distant third.
Arguments that fail numbers 1 and 2 and rest entirely on number 3 for support are not examples of science. Many times we have debated individuals with a vested interest in animal experimentation who gloss over the first two tenets of respectability, then state inflexibly that animal experiments have been used to cure human disease, as follows:
- We wish to prove that animal experimentation can lead to cures for human disease
- We state, as an authority on animal experimentation, that it in fact does lead to cures for human disease
- Therefore: we have "proved" that experiments on animals lead to cures for human disease.
This is classic fallacious reasoning.
They provide no theoretical basis for why this should be true, nor do they provide data, from observation or experiment to prove their position. The argument that animal experiments do or do not lead to cures for human disease cannot be made without the first two criteria. Mere statements that fulfill criteria number 3 are insufficient. Arguments that favor animal experiments usually rely on number 3.
By contrast, the theory that animals are not reliable models for human disease possesses respectability on all three hierarchical levels. 1. There exists a theory that animals do not reliably model human disease (which we will explore presently), 2. Volumes of data support the opinion that it does not work, and 3. Many statements from experts confirm this.
How We Define Science
After exploring the philosophy of science, we distinguish science (as opposed to pseudoscience) by the following criteria:
- Predictability - A scientific model allows us to predict subsequent events.
- Repeatability - Other scientists can reproduce the phenomena in other labs and settings.
- Parsimony or Occam's razor - The scientific model that explains the phenomena in the simplest way has the most worth.
- Mensuration - Measurements use universally accepted scales.
- Heuristic procedure - The knowledge stimulates more investigation that confirms the knowledge.
- Generality - The greater the range of data covered by a scientific model the better.
- Consilience - The data produced conforms to known data in other fields
Hence, a theory proposing to be scientific should be evaluated based on how well it accounts for the data based on the above criteria.
Predictability is considered by many to be the most important. Predictability most readily and reliably distinguishes between science and pseudoscience. Science allows predictability. Vivisectors acknowledge this when they seek to reproduce human-based data from animals and call it new, thus fulfilling the criteria of predictability. Although animals can usually be found that will demonstrate a concept of physiology, biochemistry or anatomy that is already known from human-based study, retrospective demonstration is not predictive. It is extravagant and unnecessary.
Casual vs. Causal Relationships
One purpose of science is to establish whether a relationship is causal or casual. A causally relevant relationship is different from a casual and hence irrelevant (scientifically speaking) relationship. A causally relevant relationship would be smoking and cancer. A casual and irrelevant relationship would be backseats of automobiles and pregnancy. Women may get pregnant in backseats but backseats do not cause the pregnancy.
It is a logical fallacy to confuse the two types of relationships. Animal experimenters say that all drugs currently used have been tested on animals. That is a true statement. Just as a woman may have become pregnant in the backseat of a car, so too all drugs currently in use have been tested on animals. But the reason they are tested on animals is that the law requires them to be not because the animal tests gave useful knowledge about how the drug would affect humans. The reason the woman became pregnant was caused by the combination of the egg with the sperm, not by the backseat.
Vested interest groups suggest a causal relationship between animal experimentation and all the great discoveries of science - the decrease in infant mortality, the discovery of antibiotics and vaccines, as well as the invention of artificial joints and machines for imaging the body such as X-ray, MRI, CT, fMRI, PET scan etc. They simply state as fact that all this evolved as a result of experiments on animals. They do not provide a theory as to why such a thing could have happened, nor articles from the scientific literature (observational results or the results from controlled experiments) explaining each step of discovery in support of their conclusion. They just say A caused B. No proof is offered. This is again an example of fallacious reasoning.
Accumulation of Facts vs Science
Animal experimenters frequently disavow that they are trying to cure human disease and insist that they are simply adding facts to the world of knowledge. We do not dispute that the additions of new facts can be accomplished by experimenting on animals. However, science and the accumulation of facts are not synonymous.
Martin Curd and J. A. Cover state, "…Truth by itself cannot be sufficient as a characteristic of the goal of science. This is relevant because so many of the true statements we could make about the natural world have little or no scientific value. Imagine, for example, that a biologist wants to increase our store of scientific knowledge by counting the precise number of hairs on individual dogs at various times on various days, not to test a theory or experiment with a drug to prevent hair loss but simply to know the canine hair count for its own sake. Even if the information that the biologist collects is true, it has negligible scientific value…[By contrast] Scientists are interested…in the form of general theories and laws with predictive power. These criteria of scientific excellence - generality and predictive power - and many others besides (such as explanatory power and simplicity) are among the cognitive values of science. They are not the same as truth." [9]
Science vs. Dogma
We must also distinguish between science and dogmatic adherence to unfounded beliefs. Whereas dogmatism demands that its constituency not question the beliefs of the system, science welcomes and even initiates questioning. Followers of dogma are not to study it, nor examine its veracity, nor weigh whether alternatives better explain the system governed by the dogma. They cannot debate the fundamentals upon which the system is based. They are taught unquestioning belief, not to search for truth. Science, on the other hand, withstands questioning from every quarter. In any forum, all experts' opinions bear consideration, and that consideration will through consensus determine the present understanding of truth.
German philosopher Jurgen Habermas "stressed the importance of public debate and rational consensus for preventing the domination of society by one group of interests. Consensus suffers inaccuracy when relevant opinions are suppressed. An egregious example was the suppression of Mendelian genetics in the Soviet Union in the 1930s."[10] Likewise Nazi Germany rejected Einstein's theory of relativity because he was Jewish.[11] As long as the vested interest groups control who is and who is not allowed to speak on an issue, just like Nazi Germany and the Soviet Union, only one view will be heard. Today, AFMA is repeatedly denied influence over consensus when animal experimenters fail to participate in prearranged debates or allow us access to publishing in the scientific journals.
A scientist once said, "Anyone who wishes to think rationally should have the habit of thinking coolly, with all affective feelings or sentiments and all emotions parked outside. The heat of the passions, especially if they are strong and violent bodily commotions, cannot help but cause a disturbance or even a distortion of all intellectual work." Along the same lines, mathematician Mark Kac once said a proof is something that convinces a reasonable man and a rigorous proof convinces an unreasonable man." While we agree we must point out that unreasonable men may not be convinced regardless of the persuasiveness of the proof. The easiest way to make a man unreasonable is to make his livelihood dependant on a certain activity. The man whose livelihood is threatened by a new idea will not necessarily be reasonable, rational, nor able to think coolly.
Intent
Science also assumes honest intent. It assumes a person will not lie about the results of an experiment just to keep his job, earn a livelihood, or maintain his ego.
Why animal models fail to meet these criteria
Evolutionary Biology
Now that we have examined what science is and what it is not, we will look at evolutionary biology. Evolutionary biology lies at the heart of our argument that animal-models of human disease are scientifically untenable. Speciation is both the reason why it appears that we can use animal-models as well as the reason why in reality we cannot.
D. J. Futuyma stated, "Evolution…is the central unifying concept of Biology. By extension, it affects almost all other fields of knowledge and must be considered one of the most influential concepts in Western thought."[12] Lafollette and Shanks state in Brute Science, "Since phylogenetically related species, say mammals, have all evolved from the same ancestral species, we would expect them to be, in some respects, biologically similar. Nonetheless, evolution also leads us to expect important biological differences between species; after all, the species have adapted to different ecological niches. However, Darwin's theory does not tell us how pervasive or significant those differences will be. This again brings the ontological problem of relevance to the fore. Will the similarities between species be pervasive and deep enough to justify extrapolation from animal test subjects to humans? Or will the biological differences be quantitatively or qualitatively substantial enough to make such extrapolations scientifically dubious?"[13]
Lewis Wolpert summarizes this: "Compare one's body to that of a chimpanzee - there are many similarities. Look for example, at its arms or legs, which have rather different proportion from our own, but are basically the same. If we look at the internal organs there is not much to distinguish a chimpanzee's heart or liver from our own. Even if we examined the cells in these organs we will again find that they are very similar to ours. Yet we are different, very different from chimpanzees…We possess no cell types that the chimpanzee does not, nor does the chimpanzee have any cells that we do not have. The difference between us and the chimpanzees lies in the spatial organization of the cells." [14]
One reason for the difference between species vis-à-vis the spatial organization of the cells lies within the genes. Genes can be divided into structural and regulatory genes. The structural genes allow similarities in structure and the regulatory genes account for difference between chimpanzees and humans. King and Wilson write, "Small differences in the timing of activation or in the level of activity of a single gene could in principle influence considerably the systems controlling embryonic development. The organismal differences between chimpanzees and humans would then result chiefly from genetic changes in a few regulatory systems, while amino acid substitutions in general would rarely be a key factor in major adaptive shifts."[15] Lafollette and Shanks go on to say, understanding the role of regulatory genes in evolution is "crucial to a proper understanding of biological phenomena. First, they focus our attention not merely on structural similarities and differences between organisms but also on the similarities and differences in regulatory mechanisms. Second, they illustrate an important fact about complex, evolved animal systems: very small differences between them can be of enormous biological significance. Profound differences between species need not indicate any large quantitative genetic differences between them. Instead, even very small differences, allowed to propagate in developmental time, can have dramatic morphological and physiological consequences." (Emphasis added)[16] This is why small difference between species lead to huge differences at the cellular level which is where we focus when treating disease. This is the crux of our argument.
A more concise way of explaining this would be to say that biological organisms are examples of a nonlinear complex system and that explains why small differences between biological systems negate extrapolation. (For a far more detailed explanation see Lafollette and Shanks. Brute Science Routledge 1996, Depew, D and Weber, B. Darwinism Evolving MIT Press 1995, and Kauffman, S. Origins of Order Oxford University Press 1993.) Suffice it to say here that there are biochemical reasons for questioning the extrapolation of the results of experiments on animal to humans and that evolutionary biology supports and explains these reasons.
Causal/functional asymmetry
Early animal experimenters assumed that if a tissue in two species performs the same function - say, respiration, for example - then the causal mechanism of the function is the same. They did not know any better and up until recently, so little was known about physiology at the cellular level that the assumption appeared correct; just as Newtonian physics appeared correct. Evolutionary biology however has taught us that the same function can be arrived at by different evolutionary pathways and different causal mechanisms. Birds ventilate differently from humans - the causal mechanism is different - but accomplish the same function, breathing.
This is called causal/functional asymmetry and has major implications for extrapolating data between species. The causal/functional asymmetry theory states: "although we cannot infer similarity of causal properties from similarity of functional properties, we can infer differences in causal properties from differences in functional properties."[17]
Evolution may have ended with birds and humans both exchanging gases via the lungs but it got there in different ways for the two species. Claude Bernard and other nineteenth century animal experimenters rejected the theory of evolution upon which modern biology is based.[18] They did not acknowledge the differences that speciation has introduced. Bernard's modern-day followers deny evolutionary truth every time they conduct an animal experiment for the purpose of learning about human disease.
Animal Models
Animals are used in research as models of humans. The term model in this usage denotes not "a small version of the thing itself" nor "a blueprint or design of the thing itself." A model here is a device that enables us to conceptualize unfamiliar phenomena by analogy to qualitatively different but familiar phenomena.[19]
Nonetheless, vested interest groups make extravagant claims for animal experimentation and animal models of human disease:
Every major medical advance of this century has depended on animal research.[20]
In truth there are no basic differences between the physiology of laboratory animals and humans.[21]
…we can not think of an area of medical research that does not owe many of its most important advances to animal experiments.[22]
Virtually all medical knowledge and treatment - certainly almost every medical breakthrough of the last century - has involved research with animals. There is a compelling reason for using animals in research. The reason is that we have no other choice…There are no alternatives to animal research.[23]
Virtually every major medical advance of the last 100 years (as well as advances in veterinary medicine) has depended on research with animals. Animal studies have provided the scientific knowledge that allows health care providers to improve the quality of life for humans and animals by preventing and treating diseases and disorders, and by easing pain and suffering. Knowledge gained from animal research has contributed immeasurably to a dramatically increased human life span.[24]
...virtually every advance in medical science in the 20th century, from antibiotics and vaccines to antidepressant drugs and organ transplantation, has been achieved either directly or indirectly through the use of animals in laboratory experiments.[25]
…research with animals has made possible most of the advances in medicine that we today take for granted...[26]
As we have said previously, a theory or in this case, a model is reliable or scientific, if it has predictive value. If nonhuman animals responded the same way as humans do to medications, surgery or environmental influences, then that would be evidence that the animal model is a good scientific paradigm.
Researchers maintain that animals are causal analogical models (CAMs) and can be used to study human disease. Causal analogies are a subset of analogy arguments in which causal assumptions arise based on the model. LaFollette and Shanks explain that the first condition that must be met in order for a thing to be considered a CAM is this: "X (the model) is similar to Y (the object being modeled) in respects {a…e}. X has additional property f. While f has not been observed directly in Y, likely Y also has property f."[27] So if drug Z causes death in an animal model (e.g., penicillin kills a guinea pig), animal experimenters reason by analogy that it will also cause death in humans. Animals are used as causal analogical models. And the reasoning process used is called causal analogical reasoning (CAR).
LaFollette and Shanks state that "CAMs must satisfy two further conditions: (1) the common properties {a,…,e} must be causal properties which (2) are causally connected with the property {f} we wish to project - specifically, {f} should stand as the cause(s) or effect(s) of the features {a,…,e} in the model."[28]
Evidently then, by using animals as CAMs, proponents of animal experimentation allege value well beyond the experiments' heuristic value. (Heuristic means inciting subsequent investigation.) Even extraneous observation can be heuristic. One notable scientific discovery came about while the scientist was watching sailboats. In this case, the sailboats were heuristic. But we should not give billions of dollars to scientists in order for them to draw conclusions from extraneous observation.
The pervasiveness and acceptance of lab animal CAMs suggest a rigor that the experiments simply do not have. Completely isomorphic systems have a one-to-one correspondence between all elements in each system. (Isomorphic means similarity between different organisms.) No species is 100% isomorphic with another and no one seriously claims that nonhuman animals are completely isomorphic to humans. With systems as complex as the anatomy, physiology, and biochemistry of human and nonhuman animals, we now know that even infinitesimal dissimilarities are not incidental. Dissimilarities not only negate isomorphism, but can also give rise to additional differences in a nonlinear fashion. However, the question remains: Are intact animals good CAMs so they can be used to predict what a drug or procedure will do when applied to a human?
Given evolutionary biology, there are reasons to think not. The causal/functional asymmetry theory implies that causal mechanisms may differ between species. Causal disanalogies mandate caution in extrapolating data between species. However there is data supporting the use of animal models as CAMs (penicillin cures infections in mice) and data refuting the use of animal-models as CAMs (penicillin kills guinea pigs). So we will examine the issue more closely.
We agree that nonhuman animals and humans have things in common. Both are in the Animal Kingdom. Both are composed of DNA and cells, utilize ATP and propagate certain information via action potentials. Humans have up to 98 percent of the same DNA as nonhuman primates.[29]
However there are differences. Humans are the only primates that lack the glycoprotein (sugar) molecule sialic acid on the surface of their cells. This may explain why nonhuman primates are so immune to diseases like malaria, prostate cancer, and cholera.[30]
In humans HIV binds to the white blood cell via both the CCR5 and CD4 receptors on the surface. SIV, the simian version of HIV, binds to the CCR5 receptor without binding to the CD4 receptor. A single amino acid in the CCR5 terminus is responsible for this difference. Just as a single amino acid difference is responsible for the difference between the hemoglobin molecule in humans with normal blood and the hemoglobin molecule in patients with sickle cell anemia. Just as a single amino acid difference is responsible for cystic fibrosis. Very small differences on the cellular level lead to dramatic differences in the organism as a whole.
The use of animal CAMs also suffers from the systemic disanalogy argument. Since systems (organs, tissues etc.) may differ in subtle and unknown ways, the same exposures often cause different reactions in different species. In other words, for a CAM to be predictive, "there should be no causally-relevant disanalogies between the model and the thing being modeled."[31] Considering our knowledge of evolutionary biology, this is impossible without total knowledge of both the model (animal) and thing being modeled (human).
Additional problems thwart animal models effectiveness as CAMs. CAMs must resemble the subject being modeled in all of the important respects. In terms of disease, CAMs assume the same 1) symptoms, 2) postulated etiology, 3) neurobiological mechanism, and 4) treatment response. The truth is that very rarely, if ever will two species fulfill all four criteria for any given disease.
We do not know in advance which animal will simulate the medical condition in humans. We can only know that after studies in humans. Very small differences between humans and animals can lead to lethal errors when applying animal-model-based data to humans. Which animal is like the human? Compare the results of giving humans, mice and rabbits the drugs penicillin and thalidomide. Thalidomide acts on some rabbits as it does humans - causing specific birth defects. However, penicillin does not act on rabbits as it does on humans. Mice react to penicillin the same as humans but not to thalidomide. How do you know in advance which animal will simulate the human condition? The unknowns between species are ubiquitous.
To repeat, CAMs must have: 1) common causal features, 2) causal connections between the features, and 3) no causally relevant disanalogies. None of these can be known until we know 100% about the phenomena in humans. Animals can only be proven to be "models" empirically. That is to say, we must know what happens in humans first, then study animals to see if a particular animal replicates the human condition. Only by comparing results from experiments on animals with the results from human-based data can we determine if nonhuman animals are sufficiently similar to human to allow the extrapolation of experimental results as regards that particular substance or treatment only. But this is a catch-22. We can only know which animal mimics humans after we know what happens in humans. But after we know how humans respond there is no need to use animals. This gives us no new knowledge, is obviously not predictive, and thus obviates the need for animals.
Therefore, it is a logical fallacy - circular reasoning - to use animal models. Again, we cannot say that an animal is a good model until we know that it reacts to a stimulus the same way humans do. We can only do this retrospectively. Therefore, animal models cannot be predictive. Animal models cannot prove a causal relationship in humans for the previously mentioned reasons. Epidemiology, in vitro research, clinical research, autopsies and other human-based research offer results that are much more reliable and hence are far superior methods of doing research.
(It is not the purpose of this essay explore the innumerable examples of animal models giving misleading, wrong and dangerous results when applied to humans. This discussion is elsewhere.[32] We are concerned here only with supporting our conclusion against using animal models for research vis-à-vis scientific theory.)
Can animal-models be of use despite the fact that they do not qualify as CAMs?
Animal experimenters will insist that animals, notwithstanding their lack of isomorphism and inability to be CAMs are still necessary because without animals researchers could not evaluate the drug or procedure in an intact system. We agree that life processes are interdependent, that the liver influences the heart, which in turn influences the brain, which in turn influences the kidneys etc. Thus, the response of an isolated heart cell to a medication does not confirm that the intact human heart will respond as predicted by the isolated heart cell. The liver may metabolize the drug to a new chemical that is toxic to the heart while the original was not. We also concede that cell cultures, computer modeling, in vitro research, etc. cannot replace the living intact system of a human being. But the question is, can the animal model do better than the alternative methods?
Can animal models still be predictive or helpful despite not being perfect? Lets go back to previous example. Lets assume system S1 has causal mechanisms {a,b,c,d,e} and system S2 has causal mechanisms {a,b,c,x,y}. If we stimulate sub-system {a,b,c} of S1 with stimuli sf and get result rf, then we would expect to get rf from {a,b,c}of S2 as well if the animal model is viable. However, this outcome will be highly probable if and only if {a,b,c} are causally independent of {d,e} and {x,y}. In biological systems, as those who argue in favor of intact systems emphasize, almost all systems interact. We have no a priori reason to think otherwise.[33]
Now it is important to point out that not all characteristics of the model need be present in the thing being modeled. But the similarities must be causally relevant. In our example property f must be causally connected to {a…e}.
Consider the following: We have 2 books, book A and book B. Book A has a) pages, b) a cover, c) words d) an author, e) a publisher and, the unknown contents f. Book B has a) pages b) a cover c) words d) Shakespeare as the author e) publisher Z and f) all the works of Shakespeare as the contents. Can we use B as a model to predict what the contents of f are in book A? Obviously not. There must be strong causal connections between {a…e} and f in order for causal analogical reasoning to be true, causal connections that animals do not predictably demonstrate for humans as predicted by evolutionary biology.
Animal-models as a scientific paradigm
We have shown that animal models fail to meet Thagard's requirements for an endeavor to be defined as science:
1. Theory - structure, prediction, explanation, problem solving, physical foundation, etc.
The theory that animal models can give us reliable information about human disease is not founded in evolutionary biology, has no predictive power, and does not lend itself to solving problems about human disease.
2. The community of advocates of the theory. Are the practitioners in agreement on the principles of the theory and how to go about solving the problems that the theory faces? Are they trying to explain the anomalies and do they consider the anomalies important? Are they comparing the success for their theory to the success of competing theories? Is the community actively trying to prove/disprove their theory?
The practitioners of animal experimentation deny that the inability of animal-models to predict human outcomes is a problem. In fact they boast that animal models are vital to biomedical science. The vested interest groups fail to meet the "community of advocates" responsibility completely.
3. Historical context. A theory is rejected when a better one replaces it or it has failed over a long period of time to explain the anomalies and hence is considered useless. Thus, we must evaluate whether or not the theory has explained new facts and dealt with anomalies and how it stacks up against alternative theories.
Evolutionary biology has replaced the dogma under which the animal model first appeared. Other methods of studying and curing human disease have appeared - molecular biology, technology, genetics, mathematical and computer modeling and have produced predictable, reliable results, while older methods such as clinical observation, clinical research and autopsy continue to provide reliable data.
Thagard proposed that a theory reputed to be scientific be deemed pseudoscientific if and only if
1. It has been less progressive than alternative theories over a long period of time and faces many unsolved problems.
Animal models have been far less successful than the alternatives. The theory that animals can be used as models is far less successful than evolutionary biology that says they cannot.
2. The community of practitioners makes little attempt to develop the theory towards solutions of the problems, shows no concern for attempts to evaluate the theory in relation to others, and is selective in considering confirmations and disconfirmations.
The animal experimentation community, because of financial reasons has made no effort to learn new scientific research methods and replace the animal model. The non-vested interest groups (with regards to the animal model) and individuals are the ones actually doing research utilizing the alternatives to the animal model.
Animal models also fail the test of predictability vs. verifiability. Animal models can usually be found that replicate human data but verifiability is not the same as predictability.
Animal-models also fail to fulfill any of the criteria of respectability in science:
1. The study of and verification by scientific theory is most respected.
a. Evolutionary biology, the underlying theory of modern-day biology and biomedical science fails to support the use of animal-models.
2. Examples from observation and controlled experiments that refute or support the theory are second.
a. We have not listed examples of #2, as they would make this essay into a book. For examples we recommend Sacred Cows and Golden Geese by Greek and Greek. (See suggested reading)
3. Statements made by experts are a distant third.
a. Many researchers whose livelihood depends on animal models have stated that animal models are excellent CAMs. However many others have stated that they are not. We believe that statements made by a party with a vested interest that are against that interest are more reliable than others made in favor of it. It only took one memo on nicotine addiction to bring down the tobacco industry. Again, we have not included these statements by experts refuting animal models in part because of the volumes of material. The reader is again referred to Sacred Cows and Golden Geese or the vested interest section at www.curedisease.com.
Animal-models fail to fulfill Kuhn's requirements of a good scientific theory:
1. It is accurate.
a. Animal-models frequently predict outcomes totally opposite to what actually happens in humans. Most often they simply misforecast the human outcome. Occasionally an animal will mimic the human condition. Astrology will occasionally accurately predict what your day will be like but that is not science; it is luck.
2. It is both internally consistent and consistent with other knowledge of the time.
a. It was consistent with 19th century knowledge but not 21st and it certainly is not consistent with human data e.g. smoking was thought noncarcinogenic based on animal-models.
3. It should have a broad scope. That is it should have implications for things beyond that which it was originally designed to explain.
a. The results of experiments on animals vary even within strains and between sexes of the same species.
4. It should be simple
a. The concept that animals are furry-looking humans - on the molecular level - is simple. But it is wrong
5. It should be fruitful, that is it should yield new knowledge.
a. It does add knowledge about the specific animal studied; just not about human disease.
CAMs fail the test of accumulation facts vs science. Yes CAMs accumulate facts, but not facts that are relevant to human disease. It fails the science vs. dogma test as those with a vested interest in animal experimentation usually refuse to even discuss this topic in a public forum. It also fails the falsifiability test, as there is no animal-model-based outcome or outcomes that will defeat the paradigm in the view of its practitioners. The fact that smoking was not thought carcinogenic secondary to animal experiments, that penicillin kills guineas pigs, and the thousands of other times animal-models were wrong is insufficient to challenge the paradigm according to it's supporters. Many times we have debated researchers with a vested interest in animal experimentation that state that there is nothing that would convince them of the inadequacy of the paradigm. That is representative of a religion not science.
Those with a vested interest in animal-modeled research fail to distinguish between the casual relationship of animal models and medical progress and the causal relationship of medical progress and the non-animal modeled methodologies responsible for it. The fact that the practitioners of animal experimentation have a vested interest in it also makes one question whether or not they fulfill the criteria of intent.
Conclusion
Based on our examination of the philosophy of science and animal models of human disease, we conclude that animal experimentation in biomedical research is not beneficial to humans today. In short we claim that the results of experiments on animals cannot be extrapolated to humans. We claim that very small differences between nonhuman animals and humans lead to very big differences when the results of animal experiments are applied to human patients.[34] We have provided a theoretical explanation for this: causally relevant disanalogies exist between species.
We believe that the evidence presented here and in the references at the end of this essay show that when used as a predictor of knowledge about human disease, the animal-model fails as a scientific paradigm. We suggest that just as physics moved from a Newtonian paradigm to the modern physics paradigm so biomedical researchers must acknowledge that the animal-model has failed and avail themselves of the research modalities that have a scientific basis and that have in fact eased human suffering e.g., clinical observation and research, in vitro research, mathematical and computer modeling, technology, performance of autopsies, basic science research in math, physics, and chemistry, molecular biology, and genetics among others.
We are operating under the paradigm that says all animals are more similar than different. Modern evolutionary biology reveals that the differences are far more important than the similarities with regards to how the organism operates at the cellular level - the level where disease occurs. The animal model paradigm appeared viable in the 19th century when we knew so little. On the gross macroscopic level all animal were alike. Dogs had heats; so did humans. Cats had electrical activity in their brains so did humans. But today we are studying things on the very level that defines the species' as being different - the cellular level. It is unreasonable to assume that at this level what we learn about one species - the mouse, will apply to another - the human.
Just as modern physics replaced Newtonian physics without destroying Newton's law of gravity so modern evolutionary-based biomedical research can delete the animal-model without saying that animals and humans are totally different. Newtonian physics came close to the correct answer in many cases. Animal experiments gave the correct answer to questions about the very big picture of how a living organism operates. Mice and humans and dogs all have hearts that pump oxygenated blood to keep the tissues of the body alive. Modern biology gives answers too much more difficult questions that occur on a very much smaller plane of reference- the cellular level.
Newtonian physics explained some things. But it was what Newtonian physics did not explain that led to modern physics. It is time biomedical researchers recognize that based on modern-day evolutionary biology the animal-model-paradigm should not be efficacious and in fact has failed.[35]
Suggested Reading
- Sacred Cows and Golden Geese. Greek and Greek. Continuum 2000
- Brute Science. Lafollette, H and Shanks, N. Routledge 1996
- Philosophy of Science. Curd, M. and Cover, JA. Philosophy of Science Norton 1998
- Chaos. Gleick, J. Penguin Books 1987
- Logic Salmon, WC. Prentice Hall 1984
- Logic and Philosophy. Brenner, WH. Notre Dame.1993
- Informal Logic. Walton, DN. Cambridge 1989
- Aping Science. The Medical Research Modernization Committee. The Medical Research Modernization Committee has produced excellent critiques of numerous specific animal-models. Their materials can obtained through their web site www.mrmcmed.org.
- The web site for Americans For Medical Advancement can be found at www.curedisease.com
References
[1] Curd, Martin and J. A. Cover. Philosophy of Science Norton 1998 p 144
[2] Ibid. p 11-19
[3] Ibid. p 103
[4] Ibid. p 7
[5] Ibid. p 27-37
[6] Ibid. p 27-37
[7] Ibid. p 20-26
[8] as quoted in Lafollette and Shanks. Brute Science Routledge 1996 p 19
[9] Curd, Martin and J. A. Cover. Philosophy of Science Norton 1998 p 21
[10] Ibid. p244
[11] Ibid. p 212
[12] as quoted in Lafollette and Shanks. Brute Science Routledge 1996 p72
[13] Ibid. p77
[14] Ibid. p88
[15] Ibid. p89
[16] Ibid. p90
[17] Ibid. p101
[18] Ibid. p51
[19] Public Affairs Quarterly 1993;7:113-30
[20] Trull, F. Animal Models: Assessing the Scope of Their Use in Biomedical Research, Charles River, Mass.: Charles River;1987,327-36
[21] Botting and Morrison in Scientific American Feb. 1997
[22] Ibid.
[23] Animal Research Fact vs. Myth Foundation for Biomedical Research
[24] The Foundation For Biomedical Research in Animal Research and Human Health published by FBR 1992
[25] AMA White Paper 1992
[26] Sigma Xi Statements of the Use of Animals in Research, American Scientist, 80, 73-76
[27] Lafollette and Shanks. Brute Science Routledge 1996 p 63
[28] Ibid. p112
[29] New Scientist May 15, 1999 p26-30
[30] Ibid.
[31] LaFollette and Shanks. Brute Science Routledge 1996 p113
[32] Greek and Greek. Sacred Cows and Golden Geese. Continuum 2000
[33] from Lafollette and Shanks. Brute Science Routledge 1996
[34] Greek and Greek. Sacred Cows and Golden Geese. Continuum 2000
[35] LaFollette and Shanks. Brute Science Routledge 1996 p52
