Riemann, Georg Friedrich Bernhard

views updated

RIEMANN, GEORG FRIEDRICH BERNHARD

(b. Breselenz, near Dannenberg, Germany, 17 September 1826; d. Selasca, Italy, 20 July 1866).

For the original article on Riemann see DSB, vol. 11.

Whereas Georg Friedrich Riemann’s scientific results themselves have almost completely been analyzed within the original article in the DSB, this postscript emphasizes the mathematical tradition in which he stood, his reception by his contemporaries and in the time shortly after his death, and the shaping influence that he had on mathematics as a scientific discipline.

Academic Teachers Among mathematicians that influenced Riemann, perhaps the most important was Carl Friedrich Gauss. In fact, the only references in Riemann’s doctoral dissertation are to two papers of Gauss; Riemann had studied the writings of Gauss in the university libraries of Göttingen and Berlin. However, Riemann did not write his thesis under what in the early twenty-first century would be called supervision: He apparently informed Gauss about the topic of the dissertation only after he had already finished it. Still, this was the usual procedure at German universities at that time.

Moritz Abraham Stern, Riemann’s second academic teacher at Göttingen, is often termed a second-rate mathematician, sometimes even without noting that one is comparing him to Gauss. In any case, Riemann received a profound knowledge of the state of the art of analysis as taught in Germany at that time from Stern’s lectures on calculus.

More important was the influence of Peter Gustav Lejeune Dirichlet. Shortly after Riemann arrived at Berlin, Dirichlet recognized his talents and guided his studies of the literature. In particular, he introduced Riemann to the modern techniques in analysis, which had been developed in France in the beginning of the nineteenth century by mathematicians such as Augustin-Louis Cauchy, Siméon-Denis Poisson, and others, and of which many contemporary German mathematicians were unaware. This support with respect to the literature continued even when Riemann was preparing his Habilitation thesis on Fourier series at Göttingen, when Dirichlet sent him the necessary material for the introductory historical section.

One can even speculate that Ferdinand Gotthold Max Eisenstein had a decisive influence on Riemann’s mathematics, because Eisenstein found the functional equation for the L-series modulo 4 while Riemann was in Berlin, but Riemann was not interested in closer personal contact with Eisenstein. And he had other opportunities for the inspiration for the functional equation of the zeta function and its proof in his 1859 paper on the distribution of primes.

Riemann and the Dirichlet Principle Riemann’s use of the Dirichlet principle was decisive for the advancement of mathematics because of the results that he obtained by its ingenious, sometimes even bold, use in his doctoral dissertation (1851) and in his paper on Abelian functions (1857). Furthermore, it is one of the landmarks for the change from an algorithmically to a conceptually oriented view of mathematics: The function one is in search of is not given by an explicit formula but implicitly as the solution of a variational problem, even if one can deduce some of the properties that it will necessarily have, for example, that it is harmonic. Resorting to a pure existence statement was clearly ahead of the way of thinking among most mathematicians in the middle of the nineteenth century, and many of Riemann’s contemporaries felt uncomfortable with his method.

Sometimes the story of the criticism of the Dirichlet principle is depicted as a battle between the German mathematical centers at Berlin and Göttingen on the highly prestigious theory of Abelian functions in the following way: With his 1857 paper, Riemann took the lead ahead of the Berlin representative Karl Weierstrass. Weierstrass struck back after Riemann’s death by showing that variational principles need not have a solution, so that Riemann’s proofs were incomplete. But, finally, the Göttingen school under the impetus of Felix Klein vindicated Riemann’s visions.

Nevertheless, the situation was a bit more complicated: To be sure, Weierstrass read a note to the Berlin Academy in 1870 that contained a counterexample to a naive use of variational principles. In the terminology of fairy tales, however, Weierstrass was not the stepmother who poisoned Snow White but rather the child who openly said what (almost) everybody knew about the emperor’s clothes. In fact, rumors against the liberal use of variational principles had been around as early as the late 1850s and they did not come only from Weierstrass or even Berlin mathematicians—as one learns, for example, from the notes of Felice Casorati from his conversations with Leopold Kronecker in 1864 and from a letter of the Russian mathematician Georg August Thieme to Richard Dedekind on a visit to Riemann at Göttingen in the summer of 1862.

Additionally, because Weierstrass’s note was published only in the second volume of his Mathematische Werke in 1895, the first published counterexample to the Dirichlet principle was in an 1871 article by one of Riemann’s own students, Friedrich Emil Prym. Furthermore, Prym directly attacked the specific Dirichlet principle as it had been presented in Dirichlet’s lectures, whereas Weierstrass commented on variational principles in general. Also David Hilbert fully acknowledged the justification of Weierstrass’s criticism (which was all the easier for him because in 1901 he had contributed to the proof of the Dirichlet principle as used by Riemann). Even Riemann himself knew about the gaps in his reasoning but argued

© BETTMANN/CORBIS .

in a discussion with Weierstrass in 1859 that he had made use of Dirichlet’s principle only as an easy resort that was just at hand. Even if that tool was faulty, his existence theorems still were true.

In fact, not the least influence that Riemann had on mathematics was that he set out a research agenda that was pursued by other mathematicians, most prominently from the Göttingen school: It took more than half a century before Hermann Weyl would be able to transfer Riemann’s vision of a “Fläche” (which came to be known as “Riemann surface”) into a detailed definition in the language of set theory. At about the same time Richard Courant in his doctoral dissertation also completely vindicated Riemann’s use of the Dirichlet principle. But the Berlin school around Weierstrass was also active in this research beginning with Weierstrass himself, who studied Riemann’s writings and tried to translate the results into his more algorithmically oriented way of thinking even if he had problems with the Riemann’s way of reasoning. Furthermore, Weierstrass’s favorite disciple, Hermann Amandus Schwarz, proved explicit formulas that solve the mapping problem of Riemann’s doctoral dissertation for the case of polygons and also attacked the general case.

The Distribution of Riemann’s Idea Riemann exerted his influence mainly through written sources: his own publications; books by his students, themselves prominent, such as Carl Neumann’s Vorlesungen über Riemann’s Theorie der Abel'schen Integrale; and also by handwritten copies of the notes taken during his lectures. These were distributed throughout Europe, in Germany, of course, but to a great extent also in Italy and even in Russia with the Italian mathematicians being extremely receptive to Riemann’s ideas. This does not mean that his mathematical standing was not well regarded elsewhere: He was a member of the Gesellschaft der Wissenschaften in Göttingen, of the Preussische Akademie der Wissenschaften (Berlin), of the Bayerische Akademie der Wissenschaften (Munich), of the Académie des Sciences (Paris), and of the Royal Society (London).

Attendance at his lectures personally would have influenced only few mathematicians, the maximal number of thirteen students at a lecture being documented in the Göttingen archives (whereas a few years later Ernst Eduard Kummer (1810–1893) and Weierstrass would have an auditorium of about two hundred students at Berlin).

One of the results that was attributed by hearsay to Riemann is the example ?_{n = 1}^?(sin n2x) / n2 of a continuous but nondifferentiable function. Neither this example nor any concrete statement about it can be found in Riemann’s writings. Although the function is not differentiable at the points of a dense set, there are some points at which it is differentiable, and so it is not nowhere-differentiable. A complete analysis of its differentiability was only given around 1970. What the sources do reveal, however, is that both Kronecker and Weierstrass were interested in the function, that Weierstrass claimed to have a proof that this function is not differentiable at the points of a dense set, and that Riemann himself had studied the boundary behavior of theta functions to such an extent that only an interchange of limit processes was necessary to obtain this result. (In fact, from his lecture notes one learns that Riemann did not care too much about such technical questions.)

Development of Mathematics Riemann’s results and mathematical ideas have had far-reaching consequences, even on the present view of the universe: In his Habilitation lecture on the hypotheses on which geometry is founded Riemann had defined the (differential) geometric structure of space by means of a positive definite differential form, which locally induces a Euclidean structure. Already Hermann Minkowski had studied the generalization to a no longer positive definite but still nondegenerate differential form. So for his theory of general relativity Albert Einstein only had to generalize Riemann’s concept from a locally Euclidean to a locally Minkowskian structure. (In fact, Einstein had studied Riemann’s Habilitation lecture as early as in his student days.)

If one considers mathematics as a scientific discipline, Riemann’s influence on the way in which mathematical objects are conceived in the early 2000s is perhaps even more important. Before his doctoral thesis a function of a complex variable was given as an analytic term that could be used to calculate the values of this function. (Even after the thesis Riemann’s older but longer living contemporary Weierstrass followed this approach in the form of power series both successfully and influentially.) Riemann, by contrast, would rather look for a characterization of such a function by its properties, in this case, the Cauchy-Riemann differential equations. As another example, he studied the hypergeometric function not mainly by means of the series by which Gauss had defined it but by means of the differential equation that it fulfills, which would spare him long and tedious calculations. Still more impressive is the advantage that Riemann’s approach had in the theory of elliptic and Abelian functions when one compares his results with the pages of long lists of formulas that are typical for the writings of Niels Henrik Abel and, especially, Carl Gustav Jacob Jacobi.

Minkowski attributed to Dirichlet a second mathematical principle besides the one mentioned above, namely to minimize blind calculation and to maximize thoughts led by visions. His close friend Hilbert even brought forth the principle that one should lead proofs by thoughts and not by calculations, in direct connection to Riemann. In this respect, the latter was a turning point in the history of mathematics, one of the lesser examples being the fact that he was the first mathematician who not only defined the notion of an integral but also explicitly defined what it means that a function is integrable. (It is worth noting that Dedekind, Riemann’s fellow student and colleague at Göttingen, also deeply influenced by Dirichlet, approached algebra in much the same way as Riemann approached analysis and geometry. And from Dedekind a line of influence runs via Emmy Noether, Emil Artin, and Bartel Leendert van der Waerden to the structuralistic Bourbaki school of the second half of the twentieth century.)

One should stress, however, that Riemann himself was by no means averse to or even afraid of calculations: One learns from the notes of his introductory lectures on complex analysis that he would not hesitate to use besides the nonconstructive Dirichlet principle the more algorithmic and “hands-on” means of power series expansions and even the idea of analytic continuation. Furthermore, his statement that it is probable (“wahrscheinlich”) that all zeros of the zeta function have real part 1/2 was based on long explicit calculations of zeros.

This famous Riemann hypothesis was still open in 2007, even though in 2000 the Clay Foundation offered a million-dollar prize for its solution. So Riemann not only opened new, far-leading doors in science, in particular mathematics, but his work continued to show the way to new frontiers.

SUPPLEMENTARY BIBLIOGRAPHY

The 1990 edition of Riemann’s collected works includes an overview of the secondary literature; see the bibliographies in Gesammelte mathematische Werke, pp. 869–895 and 896–910.

WORKS BY RIEMANN

Narasimhan, Raghavan, ed. Gesammelte mathematische Werke, wissenschaftlicher Nachlass und Nachträge: Collected Papers. Berlin: Springer; Leipzig: Teubner, 1990. Third edition of Riemann’s collected works, originally edited by Heinrich Weber and Richard Dedekind.

Neuenschwander, Erwin. Riemann’s Einführung in die Funktionentheorie: Eine quellenkritische Edition seiner Vorlesungen mit einer Bibliographie zur Wirkungsgeschichte der Riemannschen Funktionentheorie. Göttingen: Vandenhoeck & Ruprecht, 1996. This book contains Riemann’s notes for his introductory lectures on the theory of functions of a complex variable and a bibliography of the secondary literature on Riemann.

Elstrodt, Jürgen, and Peter Ullrich. “A Real Sheet of Complex Riemannian Function Theory: A Recently Discovered Sketch by Riemann’s Own Hand.” Historia Mathematica 26 (1999): 268–288. Autographic writing of Riemann in which he gives a tour de force through his theory of complex functions, edited with commentary.

OTHER SOURCES

Laugwitz, Detlef. Bernhard Riemann, 1826–1866: Turning Points in the Conception of Mathematics. Translated by Abe Shenitzer, with the editorial assistance of the author, Hardy Grant, and Sarah Shenitzer. Basel: Birkhäuser, 1999. First published in 1996. At present the best and most extensive presentation of the life and work of Riemann.

Narasimhan, Raghavan. “Editor’s Preface (Together with a Mathematical Commentary on Some of Riemann’s Work).” In Riemann, Gesammelte athematische Werke, pp. 1–20. Berlin: Springer; Leipzig: Teubner, 1990.

Neuenschwander, Erwin. “Der Nachlass von Casorati (1835–1890) in Pavia.” Archive for History of Exact Sciences 19 (1978): 1–89.

———. “Über die Wechselwirkungen zwischen der französischen Schule, Riemann und Weierstrass. Eine Übersicht mit zwei Quellenstudien.” Archive for History of Exact Sciences 24 (1981): 221–255.

Ullrich, Peter. “Anmerkungen zum ‘Riemannschen Beispiel’ ?_{n = 1}^?(sin n2x) / n2 einer stetigen, nicht differenzierbaren Funktion.” Results in Mathematics, Resultate der Mathematik 31 (1997): 245–265.

Weil, André. “On Eisenstein’s Copy of the Disquisitiones.” In Algebraic Number Theory in Honor of K. Iwasawa, edited by J. Coates et al. Advanced Studies in Pure Mathematics 17. Boston: Academic; Tokyo: Kinokuniya, 1989.

Peter Ullrich

More From encyclopedia.com