Moore, R. 1994. The Journal of College Science Teaching, November.
The ability to write well is one of the most important skills needed for a successful career in science and other professions (see discussion in Moore 1992). Indeed, a scientist's salary, reputation, and professional development are determined largely by his or her written work--publications, grant proposals, and the like. Moreover, understanding how to write effectively can improve one's ability to think as well as to do science (Gopen and Swan 1990; Moore 1992). The importance of writing in science underlies many "scientific writing" courses and the increased use of "writing-to-learn" programs in science departments--for example, in writing-across-the-curriculum programs.
Many science teachers require their students to read research papers as a way of teaching the students about science and scientific writing (see, e.g., Greene 1991). Despite the widespread use of such assignments, there are few resources available to help teachers accomplish the intent of the assignments. Indeed, little attention has been given to particular cases of scientific discourse (Halloran 1984 and references therein); the arrangement of scientific arguments is seldom mentioned. Consequently, few teachers understand how great scientists have used rhetoric to produce the brilliant arguments that created the "truths" described in their papers.
This paper provides a scientific and rhetorical analysis of one of the most important scientific papers ever written: the paper by James Watson and Francis Crick, published in Nature, that sketched the double helical structure they had deduced for DNA (Watson and Crick 1953a). I hope to persuade readers that the tremendous impact of Watson and Crick's paper was due largely to their understanding of rhetoric and persuasion, and that their acknowledgment that a scientist's primary task is not necessarily to present "facts," but rather to make an effeetive argument. In doing so, I do not deny the importance of the technical details of Watson and Crick's model; it would not have survived had scien tists not been able to use it to predict experimental observations. However, Watson and Crick's brilliant presentation and arguments greatly enhanced the acceptance and impact of their paper. I also hope to show readers that Watson and Crick's paper- -the paper that Gunther Stent (1980) says gave birth to the science of molecular biology--is an ideal paper for teaching stu dents about science and effective scientific writing. Indeed, Watson and Crick's paper has become a model of scientific rhetoric (Halloran 1984).
Although little has been written about the rhetoric of Watson and Crick's paper, much has been written about Watson and Crick's work. The most famous of these analyses is Watson's abrasive The Double Helix, which makes explicit what is implicit in the Nature paper. The Doub1e Helix demolished the traditional view of science as an autonomous exercise of pure reason done by disembodied, selfless spirits, inexorably moving toward a true understanding of nature. Watson's book, a best-seller still used in biology classes (see discussion in Carter and Mayer 1988), greatly upset many scientists with its portrayal of scientists as avaricious creatures motivated by fame and riches, and of science as an intensely personal enterprise that suffers from human frailties. Although many scientists denounced Watson's expose as misleading, much evidence suggests otherwise. Indeed, the many priority disputes (for example, discovery of AIDS virus), heated arguments that occur at professional meetings, and the increasing incidence of fraud in science attest strongly to the personal nature of science and the all-too-human failings and biases of scientists.
Knowledge of DNA. Much was known about DNA when Watson and Crick--two equally unknown scientists--wrote their paper:
Biologists agreed that DNA is the substance that transmits genetic information. The work of Avery et al. showed this in 1944. Alfred Hershey and Martha Chase proclaimed this more boldly in 1952 when they showed that only the DNA of a bacterial virus-- and not its protein portion--must enter a host bacterium to initiate infection.
The chemical composition of DNA had been known since the 1930s. Biologists knew that DNA is made of nucleotides consisting either of a purine (adenine and guanine) or pyrimidine (thymine and cytosine) attached to a sugar (deoxyribose) and phosphate. The bonding of the sugar (with its attached base) to the phos phate formed the backbone of the molecule. Erwin Chargaff had re ported that the amount of adenine always equals that of thymine, and that the amount of guanine always equals that of cytosine (Chargaff 1950, 1951).
One or more chains ran the length of the molecule. Substructures repeated every 3.4 A and the entire structure repeated every 34 A (that is, there were about 10 bases per turn of the helix). The diameter of the molecule was 20 A.
Competing scientists. Watson and Crick competed with Linus Pauling (in California) and Maurice Wilkins and Rosalind Franklin (in London) to discover DNA's structure. There was a vague sense in England that the problem belonged to Wilkins and Franklin; after all, Watson and Crick had been ordered to study something else.
Importance of the problem. Although the structure of DNA was regarded as an important research problem, no one knew beforehand just how important it would turn out to be. Few suspected that genetic information would be transmitted by a straightforward mechanical process, or that knowing the structure of DNA would suggest how to manipulate inheritance (Halloran 1 984) .
Watson and Crick's problem was how to best use their knowledge of DNA to frame the problem rhetorically when offering their solution; that is, how to propose a structure of DNA that would account for the function of DNA. To do this, they knew they had to answer several questions:
Is DNA helical? Although most scientists thought so, Rosalind Franklin thought that the "A"-form of DNA was not helical.
If so, how many helices? Most scientists thought there were more than one. Some models (e.g., those proposed by Fraser and Pauling & Corey; see below) had three helices, as did an aborted model proposed by Watson and Crick.
What is the geometry of the sugarphosphate backbone and its attached bases? Franklin suspected that the bases were on the inside of the molecule, while Pauling and Corey's model (as well as an original model proposed by Watson and Crick) positioned the backbone on the inside with the bases sticking out.
How does the structure of DNA relate to its function?
On April 25, 1953, a young Watson and a not-quite-so-young Crick published a paper entitled "Molecular Structure of Nucleic Acids: A Structure for Deoxyribose Nucleic Acid" (Watson and Crick 1953a) that answered these questions and, within a decade, forever changed biology. Watson and Crick's 900-word paper was one of a trilogy published simultaneously in Nature (Watson and Crick 1953a, Wilkins, Stokes, and Wilson 1953, Franklin and Gosling 1953) and was the first published announcement of their geometric model claiming to depict the structure of DNA.
Watson and Crick knew that because "truth" in science is a subjective product of argument, they could not merely present "data." Rather, they had to convince readers that their claims were not only accurate, but that they were also important. Conse quently, it is not surprising that the style and substance of Watson and Crick's paper support its persuasive intent. Indeed, their enormously persuasive paper is a remarkable example of the importance of rhetoric and effective writing to effective science. More clearly than any other paper in recent years, Watson and Crick's paper stands near, if not precisely at, the center of what Thomas Kuhn (1970) would call a scientific revolution (see discussion in Halloran 1984).
Watson knew that the timing and rhetoric of their paper would greatly affect its acceptance and impact. They had already announced a structure of DNA that proved to be incorrect. This slowed their progress because they were told soon thereafter to stop studying DNA. Having once jumped the gun, Watson and Crick lavished great care on writing their Nature paper. As Watson wrote later in The Double Helix, "By the time I was back in Copenhagen, the journal containing Linus' article had arrived from the States. I quickly read it and immediately reread it. Most of the language was above me, and so I could only get a general impression of his argument. I had no way of judging whether it it made sense. The only thing I was sure of was that it was written with style. A few days later the next issue of the journal arrived, this time containing seven more Pauling articles. Again the language was dazzling and full of rhe torical tricks. One article started with the phrase, "Collagen is a very interesting protein." It inspired me to compose opening lines of the paper I would write about DNA, if I solved its structure. A sentence like "Genes are interesting to geneticists" would distinguish my way of thought from Pauling's."
Watson wanted to achieve Pauling's status--he described Pauling as "the greatest of all chemists"--yet he also wanted to be different. This presented Watson with a classical rhetorical problem: how to be the same but different. Watson knew that the only solution to this dilemma was rhetorical.
Although Watson greatly admired and enjoyed Pauling's writing--he described it as "dazzling and full of rhetorical tricks"--Watson and Crick didn't start their paper as Watson had fantasized. Rather, they began their paper with a straightforward, understated paragraph that set the task: " We wish to suggest a structure for the salt of deoxyribose nucleic acid (D.N.A.). This structure has novel features which are of considerable biological interest."
The opening sentence of this paragraph directs readers to the important scientific exigence: the structure of DNA. The second sentence ("This structure has novel...") intimates that Watson and Crick have a solution to the exigence. This rhetorical understatement not only heightens the claims of importance, but frees Watson and Crick from having to specify what about those "novel features" is "of con siderable biological interest." Readers are free to generate any solution to any biological question that they infer from the structure, with the clear implication that all has been anticipated by Watson and Crick.
The second sentence also puts Watson and Crick's chemical claim in the context of biological knowledge. Although this added context is the great importance of the paper, Watson and Crick wait until the last substantive paragraph of the paper to tell readers why their proposed structure of DNA is "of considerable biological interest": "It has not escaped our notice that the specific pairing we have postulated immediately suggests a possible copying mechanism for the genetic material."
This sentence communicates the importance of Watson and Crick's paper. Although the paper describes the structure of DNA, the consequences and import of the paper are functional: Watson and Crick show readers that by understanding the shape of things, we can better understand how things happen.
Watson and Crick knew that the marketplace for scientific ideas is crowded and that most papers are ignored. To overcome these concerns, Watson and Crick worked hard on the rhetoric of their paper so that it would have a maximum impact on their col leagues. They rejected many of the conventions of "scientific writing" to strengthen their arguments and increase the paper's impact. For example, unlike most scientists, Watson and Crick avoided statements such as "The data suggest..." and "This paper ar gues..." that deny human activities and imply that suggesting, arguing, and demonstrating--all of which are rhetorical acts--can be accomplished without human volition.
Similarly, Watson and Crick used the first person and active voice to personalize and strengthen their writing, especially when they championed their own ideas. Indeed, the paper opens with a strong statement written in active voice that sets the goal of the paper ("We wish to suggest..."). This and all of Watson and Crick's other uses of the first person put these scientists and their actions on equal terms with the objects they were studying, helping Watson and Crick show how they constructed, criticized, and manipulated claims about DNA to understand DNA. In paragraphs 1 and 4 Watson and Crick make statements; in paragraph 4 they make assumptions; in paragraph 2 they criticize statements; in paragraphs 11 and 12, they place knowledge within the framework of other knowledge.
These uses of first person (for example, "We wish to suggest...," "In our opinion...," "We believe...," "We wish to put forward...," "It has not escaped our notice...," "We have postulated...") dramatized Watson and Crick as they claimed the arguments and model as their own. In fact, just five weeks after publishing their first paper (Watson and Crick 1953a), Watson and Crick published another paper in Nature in which they referred to their model for DNA as "our model" (Watson and Crick 1953b). These uses of first person also describe how Watson and Crick think about what claims they should make about DNA, thereby impressing on readers the importance of human action as a means of doing and understanding science. This contrasts sharply with the agentless, detached image portrayed by the excessive and clumsy use of passive voice, the preferred--and ineffective--voice of most scientists (Bostian and Thering 1987i Moore 1991; Bostian 1983; Fisher-LaMay 1987). One only has to read the earlier paper by Avery et al. ( 1944i see below) or the accompanying paper of Wilkins et al. (1953) and their detached, impersonal prose ("The pur pose of this communication is...," "It may be shown that...," "It must be decided whether...," "The...significance of a two-chain nucleic acid unit has been shown...") to appreciate the impact of Watson and Crick's first-person style of writing.
The opening paragraph of Watson and Crick's paper proposed a structure for DNA that fit a chemical "slot" and a function of DNA that fit a biological "slot." To ensure that their model filled these slots, Watson and Crick immediately tried to create what Foucault (1973) calls "rhetorical space" for their model by considering and rejecting other models that competed for these slots. Paragraphs two and three do this effectively, but differ.
Watson andCrick delicately dismiss the model of Pauling and Corey as impossible based on el ementary chemistrv (that is, atoms in the core of the molecule were packed too close together; the hydrogen bonds holding the triple helix together did not contain the necessary binding forces). However, Watson and Crick do not present the exact calculations of their criticism. Rather, they assume that their criticism will be accepted and understood by their audience. Of course, the "delicate" rejection of Pauling and Corey's model (exemplified by phrases such as "in our opinion," "we believe it is not clear," "appear to be") is a conscious fabrication; in fact, Watson and Crick were astonished and jubilant to see Pauling and Corey's blunder.
Watson and Crick dismiss the model of Fraser as being "rather ill-defined." This implies that the model is not consequential enough, that it is too ambiguous for meaningful discussion or interpretation relative to measurable aspects of DNA. Actually, Watson and Crick mentioned Fraser at the insistence of Wilkins and Franklin (Fraser worked in their lab) because he had considered hydrogen-bonded bases--albeit incorrectly--before the work of Watson and Crick (Watson 1980). In this sense, the third para graph represents a compromise regarding the priority of important claims of knowledge.
These paragraphs also show the interpretative nature of the problem and the theoretical nature of Watson and Crick's solution: they argue that there are no models available to solve the scientific exigence in ways that are consistent with data. In doing so, Watson and Crick invoked and addressed conjectural aspects of interpretation, thereby re-emphasizing the importance of the problem while simultaneously creating conceptual space for their proposed solution. Specifically, Watson and Crick question critical details of Pauling's and Corey's model, yet lament the absence of such details in Fraser's model.
Although Watson and Crick's model was vulnerable to these same criticisms (that is, it fails to present exact calculations), their rhetoric created room for their model. Watson and Crick persuaded readers to reject the competing models without any supporting calculations because Watson and Crick understood that their audience knew enough biochemistry to accept their assumptions. That is, Watson and Crick assumed that readers would realize that the competing models did not match the theory believed to accurately describe nature. In 1974, Crick described the omissions in the Watson and Crick (1953a) paper as "...striking. The structure is pro duced like a rabbit out of a hat, withno indication as to how we arrived at it."
Having rejected both competing models as being either ambiguous or inconsistent with biochemical knowledge, Watson and Crick were then set to fill the void that they'd created. Here's how they did it: Paragraphs 4-8 declare Watson and Crick's proposed "novel" and "radically different structure" of DNA. These five paragraphs cast Watson and Crick's remarkably symmetrical model into words, add details, "make the usual chemical assumptions," and clarify parts of the model by referring to acceDted causal statements, other models, and priorwork. The paper's lone fig ure, to the left of paragraph four--a "purely diagrammatic" representation of two helices wound around a central axis--shows the geo metric essence of their claimed structure.
Paragraph 4 outlines the model and answers several questions that readers would pose:
Is DNA helical? Yes.
How many helices? Two.
Where is the backbone? On the outside of the molecule.
How are the chains related? Both chains twist around the outside of the core and in opposite directions, with the sequence of atoms in one chain matching that of the other chain in reverse. Readers are not yet told how the backbones are held together.
Paragraph 4 also presents dimensions of DNA: its diameter is 20 A, the height between stacked parallel bases (along the middle of the two chains) is 3.4 A, the height of one complete turn of each helix is 34 A, and the rotation from one nucleotide to the next in each backbone is 36ø. These mea surements are confirmed by X-ray data presented in experimental papers published in the same issue of Nature.
Paragraphs 6-8 satisfy the promise of the model having "novel features" by claiming that the two helices are hydrogen-bonded at the bases: the bases of one chain form hydrogen bonds with those on the other according to a fixed pattern (i.e., adenine bonds only with thymine, guanine bonds only with cytosine). This clari fied the model and helped solve a problem facing other scientists who were studying DNA. Watson and Crick hint at the functional significance of the model by pointing out that "if the sequence of bases on one chain is given, then the sequence on the other chain is automatically deter mined (paragraph 8).
Paragraph 9 contains evldence that con firms the Watson and Crick model and, in the process, shows the closeness of fit between the proposed model and Erwin Chargaffs data. This para graph concisely provides two reasons to support the Watson and Crick model. The model is empirically adequate because it is consistent with Chargaff's data, and the model explains Chargaf's data. The equal amounts of purine and pyrimidines (i.e., that A=T and G=C) become an inevitable consequence of Watson and Crick's model.
In The Double Helix, Chargaff is notable chiefly for his scorn of and condescension toward Watson and Crick. Consequently, it's not surprising that although Watson and Crick generally avoided the use of passive voice, they did not when referring to Chargaff ("It has been found experimentally that..."), thereby making Chargaff's contribution anonymous. Interestingly, Chargaff became an outspoken critic of Watson and Crick (for example, see Chargaff 1968,1974). In the 1974 Nature retrospective entitled "Molecular Biology Comes of Age," Chargaff recalls Watson and Crick's paper: "The tone was certainly unusual: Somehow oracular and imperious, almost decalogous. Difficulties were brushed aside in the Mr. Fix-it spirit that was to become so evident in our scientific literature." This quotation portrays how Watson and Crick's paper influenced not only what we know about biology, but also the way that ideas are pursued and the spirit in which science is done.
Paragraph 10 provides limited guidance about how the model might be applied. Watson and Crick speculate that their model does not apply to RNA because the atoms would be too close together.
Paragraph 11 presents the most persuasive argument for the empirical adequacy of the model; specifically, the model is "roughly compatible" with experimental data, but needs to be tested. The paragraph also describes evidence needed to confirm the model, some of which is included in the accompanying papers in the same issue of Nature.
Paragraph 12. Watson and Crick had promised that DNA was not just a complex molecule, however correctly described, but also that it is "of considerable biological interest." Watson and Crick use this paragraph to satisfy that promise by discussing the signifi cance of their proposed structure of DNA. Readers had anticipated this explanation since the second sentence of the opening paragraph: "This structure has novel features which are of consid erable biological interest." The cryptic, coy sentence "It has not escaped our notice...," describing what was ul timately important about Watson andCrick's model, was brief because it represented a compromise by the authors: the statement was wanted by Crick but not by Watson, who feared they could be wrong (Crick 1974; Watson 1980).
This paragraph established a claim to priority for the mechanism of genetic replication (ie., that the model is important because the structure of DNA also identifies the function of DNA) while avoiding any complex and potentially controversial details about the mechanism. Paragraph 13 promises that "full details" of the proposed model and its significance (i.e., Watson and Crick 1953a, b; 1954; Crick and Watson 1954). The technical account of the structure, including rough coordinates, was pub lished in the middle of 1954 in Proceedings of the Royal Society (Crick and Watson 1954), a journal that (Crick considered "obscure" (Crick 1974).
Paragraph 14 acknowledges prepublication criticism, access to un published evidence and ideas, and funding; these preconditions to Watson and Crick's work do not enter the arguments made in their paper.
"Truth" in science is the product of argument (see discussions in Gross 1990; Moore 1992; Locke 1992 and references therein). Understanding this, Watson and Crick presented three strong arguments for their model:
Elegance. The first argument was the model's elegance, especially the basepairing mechanism that holds the chains together (paragraphs 6-8). This argument is left entirely implicit; Watson and Crick assumed that their model would appeal to the reader's sense of theoretical elegance. Watson and Crick later make this argument ex plicit by claiming that DNA "has style" and "intrinsic beauty" (Crick 1974) and that "...a structure this pretty just had to exist" (Watson 1980). I argue that much of this perceived elegance lies in Watson and Crick's use of language (see above). If this is true, then the quality of science as science depends on the quality of writing as writing (see discussion in Locke 1992). Of course, although elegance is a strong premise for an argument, it is not an absolute principle. For example, Crick's subsequent model for a comma-free genetic code was elegant, but wrong.
Precise theoretical explanation. Watson and Crick's second argument was their precise theoretical explanation for what had previously been merely a curious observation: the ratios of adenine to thymine and guanine to cytosine (paragraph 9). The base-pairing mechanism of Watson and Crick's model changed Chargaff's data from curious to inevitable. Watson and Crick also leave this argument as entirely implicit: they state the ratios immediately after describing the basepairing mechanism, and expect the reader to make the connection.
Consistent with available evidence. Watson and Crick's final argument is that their model is consistent with available evidence. Unlike their previous arguments, this argument is explicit and carefully qualified (paragraph 11).
Watson and Crick use ironic understatement as their chief rhetorical device to make these arguments: each argument is presumed by Watson and Crick to be so persuasive that none needs additional support or emphasis. This rhetorical strategy portrayed Watson and Crick as extremely confident of their model.
Watson and Crick's paper had a tremendous impact because its three arguments did everything expected in a scientific argument. Indeed, Watson and Crick's rhetoric is prototypical of a successful scientific argument:
It reminded readers of the scientific exigence that existed concerning the structure of DNA. Solving this scientific exigence required the rhetorical exigence of convincing scientists that they had, in fact, solved it.
It framed that exigence as an interpretive problem involving ambiguity about the molecular structure of DNA: if we did not better understand the structure of DNA, our understanding of life would continue to be ambigu ous. Although Watson and Crick were not forced to use this rhetorical strategy, it was best in light of their desire to assure their claims to priority of their discovery.