The ``unknown heritage'': trace of a forgotten locus of mathematical sophistication (Q2271770)

Eine Erbmasse wird in \((x+\text{Rest}_1/n)=(2x+\text{Rest}_2/n) =\cdots=(nx+ \text{Rest}_n/n)\) zerlegt; (meistens \(n=7;10)\). ``The earliest known occurrence is in Fibonacci's Liber abbaci'' (p. 613); Verf. benennt weitere mittelalterliche Autoren, er folgert hinsichtlich Barthélemy de Roman's Compendy de la praticque des nombres, um 1467: ``\dots the sophisticated versions must have been developed before Fibonacci within an environment (located in Byzantium, Provence, or possibly in Sicily?) of which all direct traces have been lost, but whose mathematical level must have been quite advanced'' (p. 613 f.). -- 1) A starting point: Gemäß einem Tractatus algorismi von 1307 verteilt Jacopo da Firenze 81 Orangen auf neun Leute. ``A modem reader encountering a problem of this kind for the first time is usually stunned.'' (p. 615) Mittels einer Symbolfigur wird die Behandlung durch Leonhard Euler und in der modernen aktuellen Literatur angesprochen, während ``Jacopo probably did not know why his rule functioned \dots and here he only presents the complete calculation as a verification. However, the original inventor must have known why, one does not stumble on the structure in question by accident.'' (p. 616) Verf. bietet einen seinerzeit möglichen Ansatz vermittels rechteckiger Schemen an, baut den jeweils aktuellen Vorgang rekursiv auf, ausgehend von \(n^2\), \((n+1)(n-1)=n^2 -1\) Elementen usw., bekennt jedoch: ``No source or historian's discussion I have looked at contains the least hint that its author had seen this.'' (p. 616, n.4) ``This is obviously an argument of the same kind as those based on pebble counters or psephoi [=Steinchen] used in early classical arithmetic.'' (p. 616) -- 2) Leonardo Fibonacci: ``Fibonacci does not give the information that the amount which each son receives equals the total number of sons, although his explanation presupposes it.'' (p. 617) Detailliert beschreibt Verf. dessen auch algebraisches Vorgehen und stützt seine -- und Leonardo's? -- Überlegungen ausgiebig mit Formeln; so führt er beide Methoden Fibonacci's, ``the type where absolutely defined contributions are taken first, and a fraction of the remainder afterwards'' und ``the type where a fraction of what is at disposal is taken first and absolutely defined contributions afterwards'' als Spezialfälle auf, wobei auch ``the solution is found by means of the regula recta, that is, first-degree rhetorical algebra in which the unknown is referred to as a thing.'' (p. 618) -- 3) Maximos Planudes: Gemäß dem dortigen Exempel: 36; \(35=36-1=7\cdot 5\); dann \(35-35/7-2\); \(28-28/7-3\) usw. heißt es: ``Without support by either symbolic algebra or a geometric representation it is also difficult to see that the `theorem' holds for `any square number', and only the geometric diagram makes it evident that the procedure will continue in such a way that exactly nothing remains in the end.'' (p. 621); bei einer Rechtecksfrage \(a+b= c+d\); \(n\cdot a\cdot b=c\cdot d\) steht: ``\dots the vicinity and the absence of a claim that he invented it himself suggests it to have been at least traditional.'' (p. 622) -- 4) The mathematics of the full problem: Verf. stellt einen allgemeinen Formalismus für das ``unknown heritage'' vor, so daß ``one may ask how corresponding calculations could be made with the tools at hand in late Antiquity or the Middle Ages.'' (p. 626), er gesteht jedoch zu: ``anybody familiar with the operations on the square pattern might discover that'' (p. 626); die Möglichkeit ``on the condition of accepting fractional pebbles'' (p. 627) wird hierdurch mit Hilfe eines Rechtecksdiagramms veranschaulicht, doch: ``Visual' mathematics, seductive as it is in simple cases, becomes much more difficult than formal calculation as soon as intricacies arise; symbolic algebra conquered for good reasons.'' (p. 627, n.16); zusammengefaßt: ``In both cases, once we are so far it is legitimate to construct the rules from the equality of the first two shares only. This can be done by somewhat laborious but simple first-degree algebra -- Fibonacci shows one way to do it, but there are alternatives. It can also be done by means of two false positions, and probably by still other methods.'' (p. 628) -- 5) Fourteenth century abbacus writings: Hier werden authentische aktuelle Texte beschrieben und klassifiziert. -- 6) Arabic pseudo-kin handelt analog von zwei arabischen Texten, um 1190 und um 1400, mit Folgerung; ``The conclusion is that mathematicians from the Maghreb or al-Andalus had come to know about the problem type already before the Liber abbaci was written; but their use of a rule which is better adapted to the `Christian' version of the problem shows that this latter version with its unknown value of \(N\) [=Erbmasse etc.] was not derived from the `Jslamic' box-problem versions but was indeed the original form. \dots In any case, the aberrant character of the two Arabic problems are strong evidence that Fibonacci and Planudes did not get their problem from the Arabic world\dots'' (p. 631). -- 7) Provence and Barthélemy de Romans: ``No known source ever treated the genre as fully as Barthélemy de Romans' Compendy'' (p. 632) ``In general, Barthélemy prefers to present first the general principles of a matter, and afterwards the examples.'', wobei ``Barthélemy starts by distinguishing between deux manieres'' (p. 633), wie Fibonacci. Jens Høyrup bringt erneut seinen Formalismus ein; bei einem dieser Spezialfälle steht: ``Nothing is said about this rule corresponding to a practice, but that may be because the corresponding general rule has not yet been presented'' (p. 634), und bei einem anderen: ``The whole treatment of division according to progressions is made under the general heading of `two false positions', whose rule is simply stated as plus et plus, meins et meins\dots'' (p. 636). -- 8) Chuquet: ``What can be concluded is, firstly, that Chuquet knew the genre not only from the Compendy but also from elsewhere; secondly, that he was not very fond of Barthélemy's ways of transforming it into some kind of coherent theory -- as we know, he had his own ways. He actually closes the sequence by the remark: Toutes telles raisons facilement se peuent faire par la rigle des premiers [=Algebra]'' (p. 639). -- 9) The Aftermath in Italy: Nachdem ``The `unknown heritage' did not disappear after Chuquet'' (p. 640), folgen italienische Texte des 15. und 16. Jahrhunderts, in denen die hier behandelte Thematik teils keine, teils weiterhin volle Beachtung fand; erfreulich, von Kurt Vogel bereits vor über 30 Jahren angesprochen: Ein italienisches Rechenbuch aus dem 14. Jahrhundert, München 1977, S. V: ``Vom Stand der Mathematik in Italien in der Zeit zwischen Leonardo von Pisa und Luca Pacioli war \dots lange nur wenig bekannt. Erst neuerdings hat sich Gino Arrighi \dots besonders verdient gemacht.'' -- 10) Elsewhere: Im übrigen Europa wurde das ``unknown heritage'' im Druck verbreitet, gelangte auch in den Unterrichtskanon; dies zeigt ein Streifzug durch ausgewählte portugiesische, deutsche -- Johannes Widmanns Rechenbuch, Leipzig 1489, hat den Titel: Behende und hubsche Rechenung auff allen kauffmanschafft [!] (p. 642) -- und französische Literatur; ``\dots similarities suggest that the Portuguese writers draw on an Ibero-Provençal rather than Italian traditions\dots The German occurrences of the problem are more likely to be based on Italian inspiration.'' (p. 642) -- 11) Whence?: Drei Fakten, nämlich ``The general intellectual climate engendered by increased schooling and literacy at all levels; the recovery and digestion of the ancient mathematical legacy; the creation of new tools, first of all symbolic analysis.'' (p. 645) sind der Grund für die Aufnahme und relativ weite Verbreitung der hier behandelten Fragestellungen; mit ``Since the problem type appears to be unknown in the Arabic world'' (p. 646) werden allerhand Möglichkeiten angedeutet, die freilich noch abzuklären sind: wann, durch wen, wie und wodurch die ``unknown heritage'' zivilisiertes Allgemeingut wurden. -- 12) Who used pebbles?: ``The first question asked for the environment where the sophisticated versions of our problem were formulated and solved; the second one is a return to the question of the first origin of the simple version. \dots Should we then make the further inference that we are confronted, if not necessarily with a Pythagorean discovery then at least with a discovery belonging within the circuit of early Greek theoretical arithmetic?'' (p. 647) Nachdem Verf. etliche in der Antike möglichen Vorstellungen formelmäßig durchspielte, folgert er: ``All in all, the only argument in favour of a Greek theoreticians' discovery of these summation formulae is that their shape points with high certainty toward a derivation or proof based on pebbles\dots If this axiom is given up, we may conclude the other way around: that (heuristic) proofs based on pebbles were no Greek or Pythagorean invention but part of the heritage which the Greeks adopted from the cultures of the Near East\dots'' (p. 650). Ein stark gegliederter, auch in den zwölf Abschnitten deutlich unterteilter und mit Skizzen versehener Beitrag, der, reichlich auf Quellen gestützt, diesen in europäischen und asiatischen Schriften nachweisbaren Versuch -- einer wohl zunächst besonderen testamentarischen Verfügung; oder: einer Aufteilung von Kieseln? -- anspricht und großenteils vermittels eines ergiebigen modernen Formalismus aufzeigt. Hier wird formelmäßig vieles von dem aufgegriffen und belegt, was sonst nur als Übersetzungen aus allen möglichen Epochen vorliegt.