Collaboration in Science and Art

2012-07-19-Pierre_and_Marie_Curie.jpg 2012-07-19-holmes_watson.png

Pierre and Marie Curie (ca. 1905) Collection of the U.S. Library of Congress

Sherlock Holmes (r) and Dr. John H. Watson. Illustration by Sidney Paget; collection of the U.S. Library of Congress

The stereotype of the lonely scholar working monk-like in a corner of the library can be put aside. Most academics now work with computers, accessing journals, writing or revising manuscripts and emailing papers and comments back and forth. The world has become much more interactive and this has changed the nature of academic research, including authorship. In many academic disciplines, collaboration has become the norm in recent decades: in many physical and social sciences, sole authorship of research papers is now unusual. For example, in the American Economic Review, a leading economics journal, coauthorship went from less than 20% in 1958 to 75% in 2008. Percentages are even higher in the physical sciences. And the frequency of citation of coauthored scientific research also tends to be greater: thus Blaise Cronin recently wrote that "bluntly stated, teams trump soloists when it comes to scientific output and impact." But coauthorship in many arts remains rare. Recent scholarship on collaborative science has clarified its advantages and disadvantages (strengths and weaknesses). Combining this analysis with recent trends in advanced art, we predict that collaboration in visual art will become increasingly common in the near future.

2012-07-19-Rodgers_and_Hart.jpg 2012-07-19-Lennon_McCartney.jpg

Richard Rodgers and Lorenz Hart (1936), Collection of the U.S. Library of Congress

John Lennon and Paul McCartney (1964), Collection of the U.S. Library of Congress

Michael Nielsen's Reinventing Discovery (Princeton University Press, 2012) neatly explains the value of collaboration for solving scientific problems. Important innovations often synthesize ideas that were previously considered unrelated: the more unrelated the elements, the more radical the synthesis. Collaboration can provide access to more ideas, because teams can cover more knowledge, in a wider range of subjects, than individuals working alone. Even experts can consequently be helped by others with complementary knowledge, to consider ideas they weren't familiar with or hadn't recognized as relevant. Groups can have more expertise than individuals, because they can be more diverse.


Portrait of Albert Einstein in 1947, Collection of the U.S. Library of Congress

But collaborations do not succeed in many situations. A critical condition for their success is the existence of a shared body of knowledge and techniques, which Nielsen calls a shared praxis. A shared praxis does not exist when there is disagreement over basic values. Such disagreements destroy collaborations, for arguments cannot be settled. Arguments can be resolved only if there is general agreement on standards for what it means for analyses or procedures to be valid or correct.

Modern science offers myriad instances of the value of collaboration. A telling example involved Albert Einstein. Einstein was an outspoken opponent of the formal organization of science, and a defender of the solitary scholar: "Only a free individual can make a discovery...Can you imagine an organization of scientists making the discoveries of Charles Darwin?" Einstein's brilliant early work, including his special theory of relativity, was done in almost complete professional isolation. Late in his life, he recalled that "I lived in solitude in the country and noticed how the monotony of a quiet life stimulates the creative mind." Yet in 1912, when Einstein was searching for a mathematical approach to general relativity, it was a friend, the mathematician Marcel Grossmann, who told him that the appropriate tool was Riemannian geometry, and the two coauthored a paper that used this to make progress toward the general theory.


Apple II computer, on display at the Musée Bolo, École Polytechnique Fédérale de Lausanne, Image courtesy Wikimedia Commons

The desirability of a shared praxis explains why collaboration does not follow the model of trade or economic exchange in which each party specializes in a distinct aspect of production. Intel makes computer chips and Apple makes computers. They cooperate in exchange, but each has a particular specialization. They accept each other's work at the point of exchange. There is often some collaboration in the specification of the item to be exchanged, but not collaboration in the actual work. Academic or artistic collaboration seems to require that the collaborators check each other's contribution for error, extraneous or irrelevant work and miscommunication. Hence, collaboration seems to substitute shared praxis for specialization. Most co-authorship appears to involve authors who share the same specialization within their discipline. The necessity of shared praxis suggests that conceptual thinkers will generally collaborate with other conceptualists and experimentalists with experimentalists. Because the conceptualists and experimentalists think so differently and approach problems from different directions, it is unlikely that they will share sufficient values to make collaboration effective. It is worth stressing this result. In thinking about collaboration in general, an economist might reason that the gains would be greatest to the most diverse combinations - individuals whose knowledge and expertise covered the widest range. This would presumably involve collaboration between experimental and conceptual scholars. Yet that this rarely occurs is probably a consequence of the fact that experimental and conceptual scholars rarely have a shared praxis.

Within the sciences and social sciences, collaboration is usually episodic or short lived, though there are exceptions. It is hard for collaborators to work together over very long periods of time unless some other factor such as a family relationship ties them together. Collaboration requires not only the shared praxis, but also agreement upon the subject to be explored and the strategy to be pursued. Consequently, collaborative work tends to focus on a few articles or a book rather than a lifetime agenda. Because experimental work often has a longer gestation, there may be less collaboration among experimental innovators than among conceptual innovators.

Comparison of collaboration in the sciences and in the visual arts reveals some important differences. Though collaboration is less common in the visual arts and humanities in general, the collaboration that does take place in the arts is more enduring, often life-long and inclusive of all the artistic work of collaborators. In contrast, scientific collaboration, beyond long-term employment in the same lab, is usually confined to a single project that may produce a few articles or a book. Then collaborators move on to other work and, often, other collaborators. A productive scientist who is prone to collaboration may have dozens of collaborators over a lifetime. The explanation of this difference between collaboration in the sciences and the arts may be found in the difference in what is shared in the shared praxis. In the sciences, co-authors share methods, subject interests and a basic conceptual or experimental approach. Shared praxis in the visual arts must go much deeper to include shared personal tastes, artistic values and style of expression that are a fundamental part of the artists' identities. The need for these deeper shared characteristics explains why there are fewer collaborations, that the collaborations that do exist are longer lasting, and often combined with a personal partnership. If artistic identities are rather completely shared, then joint artistic output is much easier.

To be continued: our next (joint) post will examine the history of artistic collaboration.