Seeing More Clearly in a Blurry Landscape: Acknowledging Ambiguity in Science
Confusion, complication, and fuzziness—in a word, ambiguity—are not often associated with the natural sciences. Perhaps we might expect some level of ambiguity in academic disciplines like theology, in which people seek to understand, however dimly, an infinite, transcendent Creator, or in economics, where we try to model the complex behavior of entire societies. When it comes to the natural sciences though, we have higher expectations. For example, in my introductory biology course for non-majors, the title of chapter one in the textbook I use is, “Scientific Thinking: Your Best Pathway to Understanding the World.” 1 Likewise, the phrases “scientific proof” and “scientific fact” are commonly used to imply a certain unassailable standard of evidence, suggesting that the “proved” matter is beyond doubt.
If one wants to be a philosophical nit-picker, science never finally proves anything.
Such a conception of the sciences as the source of unambiguous knowledge is problematic, however, because science does not really deal with “proof” or “truth” in the sense of things that can be demonstrated in a logically unassailable way. 2 While science may well be “our best pathway to understanding the world,” 3 it is neither an infallible pathway nor capable of leading us to enlightenment in isolation from other sources of knowledge.
Numerous controversies and debates being waged in the public sphere demonstrate all too well that certainty can be hard to come by, even in the sciences. Consider the public debate around anthropogenic climate change, or the recurring pitched battles over the place of evolutionary theory in schools, to cite just two prominent examples. Such conversations are often marked by polarization between those on one side, who count themselves as part of a broad scientific consensus, and their adversaries, usually a vocal minority. Interestingly, though the minorities may be painted as “anti-science” by the scientific mainstream, very often they view themselves as the voice of good science—purveyors of the correct interpretation of the scientific evidence. 4 The skepticism and cynicism that results from such sustained controversies is particularly painful because these issues can have major implications for many central aspects of private and public life: politics, economics, education, health, and faith.
I am increasingly convinced that part of the problem in such interactions is a failure to understand that a measure of ambiguity is an inescapable part of science. Expectations that “good” science should offer unambiguous, irrefutable answers to questions about public policy or personal faith, make productive conversation difficult. The dissenting minority cannot be swayed by scientific orthodoxy, because there is always some seed of doubt, ambiguity or incompleteness in the consensus opinion that can be taken as an indication that the whole edifice is unstable. Those in the majority wonder how anyone could be so obstinate, ignorant, irrational, or malevolent as to refuse to accept the clear scientific truth. They will have difficulty convincing the minority of this truth, however, because it is human nature to dig in one’s heels when one is labeled as ignorant or irrational. The opposing sides become entrenched and, ironically, the “scientific” debate becomes a matter of opposing opinions rather than a careful consideration of underlying assumptions, actual evidence, and relevant input from non-scientific disciplines. When such arguments hit the media, the typical concern is to ensure merely that both sides have equal airtime, as if any attempt to evaluate or arbitrate the conflicting positions would be futile or unfair.
I want to suggest that the sciences can indeed lead us toward clarity and understanding. In order for them to do so, however, we will need to acknowledge that ambiguity is a normal, inescapable aspect of science. On the one hand, failure to acknowledge ambiguity can lead to a denial of the limitations of science and a mistaken conclusion that science alone is sufficient to answer all our questions and make all our decisions. On the other hand, denying that ambiguity and science can co-exist can lead to complete rejection of a well-established scientific consensus when uncertainty does arise. And since uncertainty and ambiguity are inescapable, it becomes increasingly difficult to maintain our confidence in objective reality; we end up sliding into radical subjectivity and relativism.
AMBIGUITY AND THE SCIENTIFIC METHOD
Science is not so much a subject as it is a method for acquiring knowledge about the world. To see how ambiguity arises in science, it is helpful to consider how scientists go about their work. The scientific method, as it is commonly described, 5 involves (1) beginning with a set of observations relating to a particular pattern or phenomenon, (2) proposing a hypothesis to explain some aspect of that pattern or phenomenon, (3) gathering information to test that hypothesis, and (4) interpreting the information gathered.
It is fairly easy to see how ambiguity can arise in step three owing to uncertainties and the potential for errors in measurement. A dramatic example of this (though certainly not the only one) is the aptly named uncertainty principle in quantum mechanics. This principle was originally described by Werner Heisenberg, who demonstrated that it is not possible to measure precisely both the position and momentum of a particle—the more precisely we know one of these quantities, the less precise our knowledge of the other must be. Thus, we cannot know precisely both where a thing is and how fast it is moving. 6 Ambiguities in measurement or data collection have profound implications for the limits of scientific knowledge, but in what follows I want to take a closer look at ambiguities in the earlier and later steps of the method, which I think have not received as much attention.
AMBIGUITY IN CATEGORIZING: THE SPECIES PROBLEM
The first two steps of the method, as outlined above, can be affected by ambiguities that hinder our ability to categorize things in the natural world. Without clear categories, it is difficult even to describe or ask fruitful questions about the phenomena we would like to investigate. An excellent example of how our subjects can defy our attempts to categorize them is the so-called “species problem” in biology: how do we determine which organisms belong together as one species, and which we should classify as some other kind of thing?
The most influential approach to demarcating species from each other is the “biological species concept,” which was originally formulated by the great ecologist Ernst Mayr in 1942. Mayr’s concept defines a species as a population (or group of populations) in which the individuals are capable of interbreeding to produce viable, fertile offspring. 7 The latter caveat is important, as many species are capable of mating and even producing viable offspring, but those offspring are then infertile—they cannot perpetuate themselves and end up as genealogical dead-ends. Thus, for example, zebras (Equus quagga) and donkeys (Equus africanus) are considered different species, because although they can interbreed and produce offspring, those “zonkey” offspring are not fertile. Likewise, lions and tigers are different species because their hybrids (ligers or tigrons) are also sterile.
The biological species concept is intuitive, clear, and works well for distinguishing several different types of organisms, but it is not without problems. For example, what is one to make of a St. Bernard and a chihuahua? Do these dog breeds belong to same species? Mating between a female chihuahua and male St. Bernard, or vice versa, strikes me as physically impractical (in technical jargon, biologists refer to such obstacles to interbreeding as “mechanical barriers”). Even if two such individuals could mate, it seems unlikely that the chihuahua would be able to carry the resulting litter of puppies to term. Should we therefore consider these breeds to be separate species? Aren’t they all just “dogs”?
Perhaps we can allow for some flexibility here and accept that these two breeds belong to the same species, Canis familiaris, in that the potential remains for gene flow between them via matings with intermediate-sized dogs. By this standard, however, they should also be considered to be of the same species as the red wolf, Canis rufus (or any wolf, for that matter 8). Surviving red wolves are found only in the southeastern United States where they have been beneficiaries of a vigorous conservation program implemented by the U.S. Fish and Wildlife Service. In fact, red wolves exist in the wild today only because they have been re-introduced from captive populations established by the Fish and Wildlife Service in the 1970s. 9 However, recent genetic studies have shown that the animals we call red wolves are likely to be hybrids of gray wolves and coyotes, the latter being another species (Canis latrans) with rather fuzzy boundaries separating it from other canids. If red wolves are indeed hybrids of this sort, then according to the biological species concept, they are not a distinct species at all.
The ambiguous status of red wolves illustrates that the species problem is not merely an esoteric issue of interest only to professional mammalogists—it has real political and social implications. Are red wolves worth the cost and effort of a protection program if they aren’t a distinct species? Granted, there are several other ways of defining species besides the biological species concept, and one could argue for the protection of red wolves by using one of these alternative definitions. However, note that the key questions here are not so much scientific as philosophical. What is a species? What makes a population of organisms worth protecting? Science can certainly help us as we think about these issues, and perhaps provide us with some answers, but only if we have a clearly articulated, justifiable philosophical position to build on.
Fuzziness and flexibility in interbreeding is only one of the sources of ambiguity with respect to defining species boundaries. Returning to the biological species concept, we should note that because this concept uses sexual reproduction as its boundary-defining criterion, it cannot apply to organisms that reproduce asexually. Although we may be inclined to think of asexual reproduction as strange and rare, it is in fact a widespread mode of reproduction in all biological domains—more often the rule than the exception. For instance, all bacteria reproduce exclusively asexually, and combined, bacteria outnumber and outweigh all other living things on the planet. 10 Given the lack of sexual reproduction in bacteria, 11 some scientists have suggested that the very concept of a species may not apply to these most abundant of all organisms. 12
When it comes to categorizing living things, there is ample ambiguity—the boundaries between species are frequently blurry, and we have not yet arrived at a single definition of “species” that is suitable for all situations. Acknowledging this blurriness, however, does not mean denying any differences at all among living things. The difference between a little ambiguity and complete uncertainty ought to be an obvious point, but let me restate it anyway: the existence of some ambiguity or uncertainty about a particular process, relationship, or phenomenon does not mean that everything about that phenomenon is ambiguous. Thus, the fact that it can be difficult to distinguish certain closely related species, or even to classify large categories of living things, does not make biological classification futile or unimportant. Though it is difficult to organize the diversity of living things on the planet into neat boxes, it does not follow that biodiversity is a meaningless concept, or that biologists have nothing to say about its nature or importance. At the same time, we cannot simply ignore these ambiguities and pretend that “science will come up with the answers” to our questions about conservation, or any topic that requires clarity on the matter of classification, without our needing to do some hard philosophical work as well.
AMBIGUITY IN INTERPRETING OBSERVATIONS: HUMAN ORIGINS
In the latter step(s) of the scientific method, as we begin to draw conclusions about what we have seen, we may find that there is more than one logically possible interpretation of data. In other words, the meaning of our observations may be ambiguous as well. While we often speak of “scientific proof” of something or other, formally speaking, this is a kind of shorthand. If one wants to be a philosophical nit-picker, science never finally proves anything. For instance, perhaps all of our observations of things falling, which we think are explained by gravity, are actually due to the influence of some other force entirely, and the fact that gravitational theory has allowed us to accurately predict what will happen to a dropped object is merely due to an enormous set of coincidences. Tomorrow, we might all float away. As unlikely as that is, in principle it remains a possibility. As clear as the implication of an experimental result or observation may be, it is generally possible to propose an alternative explanation. Philosophers of science refer to this situation as an example of “underdetermination”: no set of empirical observations can unambiguously lead us to the correct explanation of those observations. 13
As an example, let me turn to the theory of evolution and confront the topic that seems most often to make my students uneasy: human origins. Were humans created directly by God in our modern form, or did we arise by gradual change over time, from an ancestor we share with the modern great apes (chimpanzees, gorillas, and orangutans)? Note that the second alternative here does not exclude the possibility of a divine Creator. I believe in God, Creator of the universe. How God created and continues to act in the world is another matter. I want to be clear that the question I am posing is not whether a Creator exists, which is not really open to scientific scrutiny, but rather how the history of the living, created world has unfolded, and is unfolding. Is the idea of human evolution an illusion, or does it really describe part of the history of the created world? 14 On this matter, science does have something to say.
Often, especially in the popular media, the question of human origins is addressed through a discussion of the fossil record. Fossils have a compelling story to tell, to be sure, but it seems to me that the general public is not sufficiently aware of the extent to which evolutionary theory draws on evidence from many fields other than paleontology. Among these fields is comparative genetics—the study of genetic similarities and differences among organisms. In this section, I will describe one finding among many from this field that shed light on human origins.
Genetic information in all living organisms is encoded in long molecules of DNA. Each DNA molecule consists of a linear chain of small subunits called nucleotides, of which there are four kinds: adenosine (A), guanosine (G), cytidine (C) and thymidine (T). These four subunits serve as letters in a code—the information they carry is determined by their order in the DNA molecule, much like the order of letters in the words on this page determine the information it communicates.
A chromosome is a single molecule of DNA. Every chromosome contains many genes, each of which carries the information needed to construct a particular molecule—usually a specific protein. Chromosomes also contain discernible regions, the nucleotide sequences of which do not carry obvious genetic information but, instead, play other important roles. Such regions are identifiable by recognizable “themes” in their respective nucleotide sequences. For instance, eukaryotic organisms (the vast domain of organisms that includes all animals, fungi, and plants, among others) have regions called telomeres at both ends of each of their chromosomes. In vertebrates (animals with backbones, including humans) telomeres are readily recognizable as repetitive stretches of the nucleotide sequence, TTAGGG. Additionally, every chromosome has an internal region called a centromere, which contains other distinctive, repetitive sequences. 15
Humans have 46 chromosomes in their cells 16—one set of 23 from each of our parents. In this respect, we differ from the great apes (chimps, gorillas and orangutans), all of which have 48 chromosomes (two sets of 24). If humans arose by an evolutionary process and are descended from a common ancestor we share with the great apes, then the account of our evolutionary history must explain this difference. One hypothesis that could explain such a pattern is that the ancestral species from which all four species descended had 24 chromosomes in a set—the trait that the great apes continue to exhibit—while this number was somehow reduced by one in the human lineage. How was the number reduced? The mechanism least likely to result in catastrophic loss of genetic information would involve end-to-end fusion of two of the 24 chromosomes inherited from the ancestor we shared with great apes. Chromosomal fusion could have reduced the number of chromosomes in a set to 23, without a corresponding loss of much genetic information.
The hypothesis of chromosomal fusion is a testable hypothesis in that it implies clear, observable predictions about what we would expect to see when comparing human chromosomes with those of our putative great ape siblings. Every chromosome in a species has distinct characteristics: a certain size, genes in a particular order, a centromere at a specific location, and a particular pattern of light and dark bands when stained and observed under a microscope. If the fusion hypothesis is true, then the human genome should include a chromosome whose size and banding pattern makes it look like two ape chromosomes stuck together. This chromosome should carry approximately the same genes found on the corresponding ape chromosomes. Assuming that the two ancestral chromosomes fused end-to-end, the resulting human chromosome should contain internal telomere sequences, derived from the fused ends of the ancestral chromosomes. Finally, it should contain the vestiges of a centromere from one of the two ancestral chromosomes, in addition to a functional, intact centromere. 17
In 1982, Yunis and Prakash reported on a detailed study of the sizes and banding patterns of chromosomes in humans, chimpanzees, gorillas and orangutans. They found that human chromosome 2 indeed appeared to be the product of a fusion that brought together two chromosomes inherited from our hominid ancestors. 18 Further characterization of chromosome 2 has revealed the presence of internal telomere sequences at the expected site of fusion, and an inactivated centromere. 19
Such observations support the hypothesis proposed above: that humans evolved from a common ancestor shared with the great apes. They do not prove that hypothesis, however; alternative explanations are possible that invoke the direct, special creation of humans by God. For example, one could propose that something about having a chromosome 2 with internal telomeres or a degenerate centromere is necessary for human life. If this is so, and if we were created by God in our current form, then the fact that our chromosome 2 appears to be the product of fusion of ancestral primate chromosomes is incidental—that’s simply the way God had to make us, in order to make humans. Alternatively, perhaps this particular configuration of chromosome 2 is not essential, but God decided to make humans with a chromosome 2 that appears as if it had been assembled by fusion of ancestral chromosomes.
Thus, the conclusion that humans have evolved from an ancestor we shared with the great apes is not an inescapable one. The data does not raise its voice and tell us how it must be interpreted. Nevertheless, the fact that logically possible alternative hypotheses can be advanced to account for human chromosome 2 does not necessarily mean that all these alternative hypotheses are equally strong. The ambiguity or underdetermination here does not mean there are no criteria for evaluating and deciding between alternative hypotheses in this case.
For a number of biological and theological reasons, the alternative hypotheses I have suggested are rather unpalatable. The first alternative, that a fused chromosome 2 is essential to human life (and therefore God must have made us this way) suggests that our humanity is contingent, at least in part, on a specific chromosomal arrangement. Biologically speaking, however, there is no a priori reason why a human would require two sets of 23 chromosomes with genetic material arranged in a particular way. Indeed, chromosomal rearrangements are common in living things. 20 While these rearrangements may not always have beneficial effects, they are not invariably fatal or otherwise transformative either. It is highly unlikely that the connection between a particular arrangement of chromosomes and human life is so rigid that we could not exist without chromosome 2 in the configuration we find it. Furthermore, it seems to me that arguments of this sort run the risk of suggesting that people with chromosomal insertions, deletions or rearrangements—this does happen—are somehow not human. Such a conclusion is at odds with biology, and theologically distasteful.
The second alternative—that God did create humans directly in our modern physical state, but made us appear as if we are modified descendants of a common primate ancestor is also troublesome. Partly to respond to this sort of theological argument, the great evolutionary geneticist Theodosius Dobzhansky published an article in 1973 entitled, “Nothing in Biology Makes Sense Except in the Light of Evolution.” 21 Dobzhansky’s main point was that the patterns we see in the genetic makeup of living things, their embryonic development, their geographic distribution, their diversity and unity, and the fossils they’ve left behind are best explained as the result of a long evolutionary history. He did not make this argument to oppose a creationist stance. Rather, he considered himself to be a creationist—one who saw evolution as the mechanism by which God creates. The notion that God made the world recently, in its current state, but shaped it to appear as if it had a long evolutionary history, was abhorrent to Dobzhansky. If this were the case, he wrote, it would imply that our Creator is capricious and deceptive, rather than consistent and trustworthy. He strongly (but rightly, I believe) dismissed that view as a form of blasphemy, contrary to the main thrust of biblical revelation and Christian tradition.
While science is not immune from ambiguity, this does not mean that our perceptions of scientific knowledge as grounded in objective reality are entirely illusory. In particular, I think it is important to take broad scientific consensuses seriously, where they are to be found. This does not mean we should always blithely accept the majority viewpoint, but neither should we dismiss a particular position simply because it is the dominant paradigm in science, or because someone can point out that the position in question is not unassailable truth. Because scientific knowledge cannot be absolutely proved, theories will always be vulnerable to claims that the science is “not settled.” Granted, stories of underdog scientists or laypeople challenging the scientific establishment, enduring ridicule and contempt, but ultimately being proved right have an undeniable appeal. However, just because mavericks are sometimes correct, we should not jump to the conclusion that the simple act of being unorthodox is a sign of reliable insight.
At the same time, we cannot afford to ignore or deny ambiguity in science. If we believe that science provides a sure antidote to ambiguity or underdetermination, we are liable to fall into the trap of thinking that science has all the answers, or at least, that science has the only answers worth seeking. In addressing the societal challenges we face, or answering pressing questions, we must pay attention to scientific information, but cannot allow it to become the only authority in our quest for clarity. 22 There are lessons to be learned and insights to be considered that natural science is simply not in a position to discover. For those we must look to philosophy, theology, economics, history, psychology, political studies, law, folk knowledge, religion, the arts, and other sources of wisdom. We must continue to value and vigorously pursue these perspectives as well if we are to bring the full range of our intellectual and moral resources to bear on the complex challenges that face us today.
- Jay Phelan. What is Life? A Guide to Biology (New York: W. H. Freeman, 2010), 1.
- A thorough justification of this statement is beyond the scope of this essay. However, interested readers will find accessible introductions to the nature of scientific reasoning and inference in Samir Okasha, Philosophy of Science: A Very Short Introduction (Oxford: Oxford University Press, 2002), 18–39, and Geoffrey Gorham, Philosophy of Science: A Beginner’s Guide (Oxford: Oneworld, 2009), 53–87.
- Depending, of course, on how we define “best,” and what constitutes “understanding.”
- An example of this sort of “persecuted minority” mindset, with respect to climate change, can be found in Lawrence Solomon, The Deniers (Minneapolis: Richard Vigilante Books, 2010). On the evolution front, think tanks like the Discovery Institute and the Institute for Creation Research carry the flag for dissenters from scientific orthodoxy, who believe that scientific data, properly interpreted, casts doubt on evolutionary theory.
- For example, see Neil A. Campbell et al., Biology, 8th ed. (San Francisco: Benjamin Cummings, 2008), 19–20. These steps may also be subdivided or amalgamated. Of course, in reality, scientific knowledge rarely accumulates in so straightforward a way. This portrayal is highly idealized. See also Sherry Seethaler, Lies, Damned Lies, and Science (Upper Saddle River, NJ: FT Press, 2009), 3–4.
- Richard P. Feynman, Easy and Not-so-Easy Pieces (London: Folio Society, 2008), 114.
- Ernst Mayr, Systematics and the Origin of Species (New York: Columbia University Press, 1942), 120–22.
- Indeed, many biologists consider domestic dogs to be a subspecies of the gray wolf, Canis lupus. See W. Christopher Wozencraft’s discussion in Mammal Species of the World: A Taxonomic and Geographic Reference, 3rd ed., s.v. “Order Carnivora.”
- Robert K. Wayne and John L. Gittleman, “The Problematic Red Wolf,” Scientific American 273 (July 1995): 36–39.
- William B. Whitman et al., “Prokaryotes: The Unseen Majority,” Proceedings of the National Academy of Sciences of the USA, 95 (1998): 6581.
- Bacteria can exchange genetic material through the process of conjugation, which is sometimes colloquially referred to as “bacterial sex.” However, conjugation differs from true sexual reproduction in a number of important ways. Most critically, it does not involve the production or fusion of gametes or genomes to generate a new individual. While one could attempt to formulate an analogue of the biological species concept, and define bacterial species on the basis of their ability to exchange genetic material with each other, such an attempt would probably generate more ambiguity than it resolves.
- Carl Zimmer, “What is a Species?” Scientific American 286 (June 2008): 79.
- Okasha, 71–76.
- Similarly, I should note that a “familial” relationship between humans and great apes need not necessarily deny our human dignity or value as having been created in God’s image. Our distant cousins may be apes, but that doesn’t mean we have to act like them.
- Despite their name, centromeres are not necessarily found at the center of the chromosome.
- There are exceptions to cells having two sets of chromosomes, but for the most part, this generalization holds.
- Centromeres play a critical role in cell division, such that chromosomes with two centromeres (dicentric chromosomes) are somewhat unstable. Hence, over time, the persistence of a dicentric chromosome in a lineage usually requires alteration of one of the centromeres. See Daniel L. Hartl’s Essential Genetics: A Genomics Perspective, 5th ed. (Boston: Jones & Bartlett, 2011), 154–55.
- Jorge T. Yunis and Om Prakash, “The Origin of Man: A Chromosomal Pictorial Legacy,” Science 215 (1982): 1525.
- Yuxin Fan et al., “Genomic structure and evolution of the ancestral chromosome fusion site in 2q13–2q14.1 and paralogous regions on other human chromosomes” Genome Research 12 (2002): 1651–62.
- Any genetics textbook will include a discussion of such chromosomal rearrangements. For one example, see Hartl, Essential Genetics, 166–77.
- Theodosius Dobzhansky, “Nothing in Biology Makes Sense Except in the Light of Evolution,” The American Biology Teacher 35 (1973): 125–29.
- For a further elaboration of this point, see Brian Wynne, “When Doubt Becomes a Weapon,” Nature 466 (2010): 441–42.