Why Be So Critical - Nineteenth Century Mathematics and The Origi
Why Be So Critical - Nineteenth Century Mathematics and The Origi
Why Be So Critical - Nineteenth Century Mathematics and The Origi
Summer 2016
Part of the Analysis Commons, Curriculum and Instruction Commons, Educational Methods
Commons, Higher Education Commons, and the Science and Mathematics Education Commons
Click here to let us know how access to this document benefits you.
Recommended Citation
Barnett, Janet Heine, "Why be so Critical? Nineteenth Century Mathematics and the Origins of Analysis"
(2016). Analysis. 1.
https://digitalcommons.ursinus.edu/triumphs_analysis/1
This Course Materials is brought to you for free and open access by the Transforming Instruction in Undergraduate
Mathematics via Primary Historical Sources (TRIUMPHS) at Digital Commons @ Ursinus College. It has been
accepted for inclusion in Analysis by an authorized administrator of Digital Commons @ Ursinus College. For more
information, please contact [email protected].
Why be so Critical?
Nineteenth Century Mathematics and the Origins of Analysis
Janet Heine Barnett∗
One striking feature of nineteenth century mathematics, as contrasted with that of previous eras,
is the higher degree of rigor and precision demanded by its practitioners. This tendency was especially
noticeable in analysis, a field of mathematics that essentially began with the “invention” of calculus
by Leibniz and Newton in the mid-17th century. Unlike the calculus studied in an undergraduate
course today, however, the calculus of Newton, Leibniz and their immediate followers focused entirely
on the study of geometric curves, using algebra (or “analysis”) as an aid in their work. This situation
changed dramatically in the 18th century when the focus of calculus shifted instead to the study of
functions, a change due largely to the influence of the Swiss mathematician and physicist Leonhard
Euler (1707–1783). In the hands of Euler and his contemporaries, functions became a powerful
problem solving and modelling tool in physics, astronomy, and related mathematical fields such as
differential equations and the calculus of variations. Why then, after nearly 200 years of success in
the development and application of calculus techniques, did 19th-century mathematicians feel the
need to bring a more critical perspective to the study of calculus? This project explores this question
through selected excerpts from the writings of the 19th century mathematicians who led the initiative
to raise the level of rigor in the field of analysis.
∗
Department of Mathematics and Physics, Colorado State University-Pueblo, Pueblo, CO 81001-4901;
[email protected].
1
∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞
1
Excerpt from Rein analytischer Beweis des Lehrsatzes, dass zwischen je zwey Werthen, die ein entgegengesetzes
Resultat gewähren, wenigstens eine reele Wurzel der Gleichung liege (Purely analytic proof of the theorem that between
any two values which give results of opposite sign there lies at least one real root of the equation). The translation of
Bolzano’s paper used in this project is taken from [Russ, 1980].
2
As was not uncommon in the nineteenth century, Bolzano’s use of the word “arithmetic” here referred to the
mathematical discipline that is today called “number theory.”
3
Excerpt from Cours d’Analyse (Course on Analysis). Translations of the two Cauchy excerpts used in this project
were prepared by the project author.
2
Augustin Cauchy, 18234
My principal aim has been to reconcile rigor, which I took as a law in my Cours d’Analyse,
with the simplicity that results from the direct consideration of infinitesimals. For this reason,
I believed I should reject the expansion of functions by infinite series whenever the series
obtained was divergent; and I found myself forced to defer Taylor’s formula until the integral
calculus, [since] this formula can not be accepted as general except when the series it rep-
resents is reduced to a finite number of terms, and completed with [a remainder given by]
a definite integral. I am aware that [Lagrange] used the formula in question as the basis of
his theory of derivative functions. However, despite the respect commanded by such a high
authority, most geometers5 now recognize the uncertainty of results to which one can be led
by the use of divergent series; and we add further that, in some cases, Taylor’s theorem seems
to furnish the expansion of a function by a convergent series, even though the sum of that
series is essentially different from the given function.
∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞
Task 1 In what way do the concerns of these three mathematicians about the relation of
calculus (analysis) to geometry, and about the state of calculus (analysis) in general,
seem to be the same/different?
4
Excerpt from Résumé des leçons sur le calcul infinitésimal (Summary of lessons on the infinitesimal calculus).
5
The meaning of the word “geometer” also changed over time; in Cauchy’s time, this word referred to any mathe-
matician (and not just someone who worked in geometry.)
6
Excerpt from Stetigkeit und irrationale Zahlen (Continuity of irrational numbers). The translation of Dedekind’s
text used in this project is taken from [Dedekind, 1901].
7
Unlike Bolzano’s use of the word “arithmetic” to mean “number theory,” Dedekind’s use of the expression “scientific
foundation for arithmetic” was related to the set of real numbers and its underlying structure.
3
Task 2 This task looks at some of the mathematical results mentioned by Bolzano, Cauchy
and Dedekind.
Task 3 Find at least two references in Abel’s letter to infinite series as an important concept
or issue in mathematics. To what degree do the concerns that Cauchy expressed about
series agree with Abel’s view of series?
Task 4 What was it that Abel thought was “exceedingly surprising” about the state of mathe-
matics at the time? Be specific here! Do you agree with his reaction to that state of
affairs? Explain.
Task 5 Towards the end of the excerpt that we are reading from his letter, Abel remarked
that a series of the following form can be convergent for “x less than 1,” but divergent
for x = 1:
ϕ(x) = a0 + a1 x + a2 x2 + . . .
(a) Provide an example in which this occurs, specifying both the series (by giving
values for the coefficients a0 , a1 , . . .) and the function ϕ(x) to which that series
converges for “x less than 1.” (This doesn’t really take much work, so don’t make
this harder than it is!)
4
(b) Notice that Abel went on to speculate that an even worse situation might occur.
Namely, he proposed the possibility that a series ϕ(x) = a0 + a1 x + a2 x2 + . . .
might be convergent for ‘x less than 1’ and convergent for x = 1, but in such a
way that lim ϕ(x) is not equal to ϕ(1).
x→1
What mathematical concept is involved here? That is, if such a function ϕ does
in fact exist, what function property is ϕ lacking?
Task 6 Consider the following series discussed by Abel at the end of this extract:
x 1 1
= sin x − sin 2x + sin 3x − etc.
2 2 3
∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞
8
The English translation of Abel’s letter used in this project is taken from [Bekken, 2003].
9
Latin for “Hold your laughter, friends!”
5
I have in general got my eyes opened in a most astonishing manner: Because when one
excludes the most simple cases, for ex. the geometric series, then in the whole of mathematics
there is almost no infinite series whose sum is determined in a strict way. In other words, the
most important part of mathematics stands there without foundation. Most of it is correct,
that is true, which is exceedingly surprising. I am working hard to search for the reason behind
this.
A very interesting task. I do not think you will be able to propose to me many theorems
in which there are infinite series, against whose proof I shall not provide reasoned objections.
Do it, and I will answer you.
[There follows a discussion, omitted here, about the Binomial Series, about
which Abel had derived certain results.]
To show by a general example how poorly one is reasoning and how careful one ought to
be, I will choose the following example: Let
a0 + a1 + a2 + a3 + a4 + etc.
be any infinite series. Then you know that a very useful way to sum this series is to search
for the sum of the following:
a0 + a1 x + a2 x2 + a3 x3 + a4 x4 + etc.
and after that to put x = 1 in the result. This may be correct, but to me it seems one
cannot assume it without proof, because even if one proves that
ϕ(x) = a0 + a1 x + a2 x2 + . . .
for all values of x less than 1, it is not because of this certain that the same thing happens
for x = 1. It could very well be possible that the series a0 + a1 x + a2 x2 + . . . approaches a
different quantity than a0 + a1 + a2 + . . . when x approaches more and more to 1. This is
clear in the general case when the series a0 + a1 + a2 + . . . is divergent, because then it has
no sum. I have proved that it is correct when the series is convergent.
The following example shows how one can cheat oneself. It can be strictly proved for all
values of x less than π that
x 1 1
= sin x − sin 2x + sin 3x − etc.
2 2 3
From this it seems to follow that the same formula should hold for x = π , but then we
would obtain . . . [an absurdity].
.........
One applies all operations to infinite series as if they were finite, but is this allowed?
Hardly! — Where is it proved that one gets the differential of an infinite series by differenti-
ating each term?
It is easy to give an example where this is not correct, for example:
x 1 1
= sin x − sin 2x + sin 3x − etc.
2 2 3
.........
∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞∞
6
3 Conclusion
The concerns expressed by Abel, Bolzano, Cauchy and Dedekind in the excerpts we have read in this
project were emblematic of the state of analysis at the turn of the nineteenth century. Ultimately,
mathematicians of that century responded to this set of concerns by moving to the requirement of
formal proof as a way to certify knowledge via the rigorous use of inequalities intended to capture the
notion of two real numbers “being close” that underlies the limit concept. Other factors that influ-
enced this direction included new teaching and research situations, such as the École Polytechnique
in Paris, that required mathematicians to think carefully about their ideas in order to explain them
to others. Today, this nineteenth-century response remains at the core of the study and practice of
real analysis. The final task in this project takes one last look at the motivations of those who led
the way in formulating this response, as they expressed it in their own words.
Task 7 Look back at the excerpts from the works of Abel, Bolzano, Cauchy and Dedekind
that we have read in this project. What questions or comments would you address to
these mathematicians about aspects of their concerns that are not addressed in the
earlier tasks? (Write at least one question and at least one comment, please!)
References
O. Bekken. The Lack of Rigour in Analysis: From Abel’s Letters and Notebooks. In O. Bekken and
R. Mosvold, editors, Study the Masters: The Abel-Fauvel Conference, pages 9–21. NCM Göteborgs
Universitet, Göteburg, 2003.
B. Bolzano. Rein analytischer Beweis des Lehrsatzes, dass zwischen je zwey Werthen, die ein
entgegengesetzes Resultat gewähren, wenigstens eine reelle Wurzel der Gleichung liege (A Purely
Analytic Proof of the Theorem that between two values which give results of opposite sign there lies
at least one real root of the equation). Gottlieb Haase, Prague, 1817. English translation by S. J.
Russ in [Russ, 1980].
A.-L Cauchy. Cours d’Analyse de L’Ecole Royale Polytechnique (Course on Analysis of the Royal
Polytechnic School). De Bure, Paris, 1821.
A.-L. Cauchy. Résumé des leçons données à L’Ecole Royale Polytechnique sur le calcul infinitésimal
(Summary of lessons given on the infinitesimal calculus at the Royal Polytechnic School). De Bure,
Paris, 1823. English translation by D. Cates in [Cates, 2012].
R. Dedekind. Stetigkeit und irrationale Zahlen (Continuity and Irrational Numbers). F. Vieweg und
Sohn, Braunschweig, 1872.
R. Dedekind. Essays on the Theory of Numbers. The Open Court Publishing Company, Chicago,
1901. Includes English translation of Dedekind’s Stetigkeit und irrationale Zahlen, translator W.
W. Beman.
E. Holst, C. Stømer, and L. Sylow, editors. Niels Henrik Abel: Memorial Publié A L’occasion
Du Centenaire De Sa Naissance (Memorial Published on the Occassion of the Centary of his
Birth). Jacob Dybwad, Kristiana, 1902. Includes full text of Abel’s correspondence in the original
Norwegian and in French translation.
7
S. B. Russ. A Translation of Bolzano’s Paper on the Intermediate Value Theorem. Historia Mathe-
matica, 5:156–185, 1980.
H. K. Sørensen. Exceptions and Counterexamples: Understanding Abel’s comment on Cauchy’s
Theorem. Historia Mathematica, 32:453–480, 2005.
8
4 Notes to Instructors
PSP Content: Topics and Goals
This Primary Source Project (PSP) is designed for use in an Introductory Analysis course. It has also
been used in History of Mathematics courses and Capstone Seminars for mathematics majors. Its
goal is to provide context for the use of rigorous proofs and precise ϵ-inequalities that developed out
of concerns about the state of analysis that first arose in the nineteenth century, but which remain
defining characteristics of today’s analysis. Both these tools of the current trade (i.e., rigorous
proof, precise inequalities) offer challenges to students of introductory analysis, who have typically
encountered calculus only as a procedural and applied discipline up to this point in their mathematical
studies. By offering a glimpse into the problems that motivated nineteenth century mathematicians
to shift towards a more formal and abstract study of the concepts underlying these procedures and
applications, the readings in this PSP provide students with a context for making a similar shift in
their own understanding of these concepts. Completing this PSP early in the course can also provide
students and instructors with a basis for reflection on and discussion of current standards of proof
and rigor throughout the course.
Student Prerequisites
The project assumes that students are familiar with fundamental concepts from a first year calculus
course, including basic results about limits and power series. However, no prior study of analysis or
experience with formal proof writing is needed.
9
Classroom Implementation Suggestions
Classroom implementation of this project can be accomplished by way of one of the two following
basic approaches; hybrids of these two methods are, of course, also possible.
• IMPLEMENTATION METHOD I
Students are assigned to read the entire PSP and respond (in writing) to the tasks therein prior
to class discussion, where students are provided with a copy of the project that leaves blank
space below each task where they can record their final responses. Typically, the author assigns
this reading one week prior to a class discussion of it; other instructors have confirmed that
sufficient time for careful advance reading is important for high quality in-class discussions.
Students are encouraged to discuss the readings and PSP tasks with each other or with the
instructor (outside of class time) before the assigned due date (provided their written responses
are their own). While there is no prohibition against using additional resources to complete
the PSP (e.g., a calculus text), it is important to assure students that there is no need to do
any historical research in order to complete it.
On the assignment due date, a whole class discussion of the reading is conducted by the
instructor, with student responses to various PSP tasks elicited during that discussion. An
instructor-prepared handout containing solutions to select tasks (especially Task 2) can be
helpful during this discussion. The completed written work is typically collected at the close
of that class period; however, the discussion could also be conducted after the instructor has
collected and read students’ written PSP work. The author does evaluate students’ individual
written work for a grade. That evaluation and grade is based primarily on completeness,
but also takes into account both presentation (e.g., use of complete sentences) and accuracy
(particularly with regard to the mathematical details in Tasks 2, 5, 6).
A brief set of “Summary Discussion Notes” that could be used by an instructor during a whole
class discussion of the PSP is offered in the Appendix to these Notes. Although some type of
summarizing discussion is highly recommended, that discussion need not adhere to the notes
provided here.
• IMPLEMENTATION METHOD II
Students are assigned to read only the primary source excerpts in the project as preparation for
small group work on project during class time. During class time, students then work together
in small groups to write their answers to the PSP tasks, with the instructor circulating between
groups to facilitate that work. The completed written work is then either collected from each
group at the close of that class period (and possibly evaluated for a grade), or students can
be asked to write formal responses to some or all of the tasks on an individual basis (again,
possibly evaluated for a grade). Instructors opting for implementation in small groups may also
wish to conduct a whole-group discussion, based on select portions of the “Summary Discussion
Notes” included below, at one or more junctures during implementation.
Depending on the course and the class period length, this implementation plan may take up to
2 full class days to complete; a sample schedule for accomplishing this is provided in the next
section of these Notes.
10
Sample Implementation Schedule (based on a 50-minute class period)
The following sample schedule, based on Implementation Method II, offers several options to help in-
structors tailor this mode of implementation to their course goals and available class time. Depending
on the exact combination of individual/small-group/whole-class work, this method of implementation
requires 1.5–2 class days (based on 50-minute class periods).
• Homework: A complete formal write-up of Tasks 2(a), 5, 6 and 7, to be due at a later date
(e.g., one week after completion of the in-class work).
10
The author’s method of ensuring that advance reading takes place is to require student completion of “Reading
Guides” (or “Entrance Tickets”) for which students receive credit for completion, but with no penalty for errors in
solutions. Students are asked to always strive to answer each question correctly, but to think of Reading Guides as
preparatory work for class, not as a final product (e.g., formal polished write-ups are not expected). Students who
arrive unprepared to discuss assignments on days when group work is conducted based on advance reading are not
allowed to participate in those groups, but are allowed to complete the in-class work independently. Guides are collected
at the end of each class period for instructor review and scoring prior to the next class period.
A typical guide will include “Classroom Preparation” exercises (generally drawn from the PSP Tasks) for students
to complete prior to arriving in class, as well as “Discussion Questions” that ask students only to read a given task and
jot down some notes in preparation for class work. Students are also encouraged to record any questions or comments
they have about the assigned reading on their guide and are sometimes explicitly prompted to write 1–3 questions or
comments about a particular primary source excerpt; their responses to such prompts are especially useful as starting
points for in-class discussions. On occasion, tasks are also assigned as follow-up to a prior class discussion.
Experience has proven the value of reproducing the full text of any assigned project task on the guide itself, with
blank space for students’ responses deliberately left below each question. This not only makes it easier for students
to jot down their thoughts as they read, but also makes their notes more readily available to them during in-class
discussions. It also makes it easier for the instructor to efficiently review each guide for completeness (or to skim
responses during class for a quick assessment of students’ understanding), and allows students to make more effective
use of their Reading Guide responses and instructor feedback on them at a later date.
11
Connections to other Primary Source Projects
The following additional projects based on primary sources are also freely available for use in an
introductory real analysis course; the PSP author name for each is listed parenthetically, along
with the project topic if this is not evident from the PSP title. Shorter PSPs that can be be
completed in at most 2 class periods are designated with an asterisk (*). Classroom-ready versions
of the last two projects listed can be downloaded from https://digitalcommons.ursinus.edu/
triumphs\_topology; all other listed projects are available at https://digitalcommons.ursinus.
edu/triumphs\_analysis.
Acknowledgments
The development of this student project has been partially supported by the TRansforming In-
struction in Undergraduate Mathematics via Primary Historical Sources (TRIUMPHS) Program
with funding from the National Science Foundation’s Improving Undergraduate STEM Education
Program under Grant No. 1523494. Any opinions, findings, and conclusions or recommendations
expressed in this project are those of the author and do not necessarily reflect the views of the
National Science Foundation.
12
With the exception of excerpts taken from published translations of
the primary sources used in this project and any direct quotes from
published secondary sources, this work is licensed under a Creative
Commons Attribution-ShareAlike 4.0 International License (https://
creativecommons.org/licenses/by-sa/4.0/legalcode). It allows
re-distribution and re-use of a licensed work on the conditions that the
creator is appropriately credited and that any derivative work is made
available under “the same, similar or a compatible license.”
13
APPENDIX: Summary Discussion Notes: Why be so Critical?
• Caution that one of the difficulties with historical readings is that the meanings of words change over time;
for example, ‘geometer’ referred to any mathematician (not just someone who worked with geometry)
• Overview of pre-nineteenth century calculus themes
Focus Primary justification of “correctness”
Time Period What objects should we study? How do we know our mathematics is “true”?
17th century Calculus of CURVES New methods produce results that matched
(using algebra as a tool) “old” (known) results (obtained from geometry)
18th century Calculus of FUNCTIONS Methods produce correct predictions
(with physics as primary motivation) (in physics)
• Overview of the situation at the end of 18th/start of 19th century (Four main points, I – IV)
I. Increasing mistrust of “geometric” intuition as valid proof method for “analytic” truths
(and more general frustration that analytic “truths” are being verified by non-analytic ‘proofs”)
Ask for evidence of this in the assigned reading.
II. Concern that existing ‘algebraic’ proof methods lack adequate rigor
Ask for evidence of this in the assigned reading; two subthemes to elicit here:
– Euclid had long been a model of rigor; nineteenth century mathematicians express desire
to bring back something like an axiomatic approach as a foundation for certain knowledge
– algebra allows too much generality (e.g., unrestricted)
Makes it too easy to assume that properties (e.g., continuity, rationality) that hold at all “lower”
values will also hold in the limit (elicit or mention Abel power series example here)
III. Use of power series (in particular) lacks firm foundation
Ask for evidence of this in the assigned reading; two mathematical points to elicit in particular:
P
– Discuss current views about ∞ n=1 x (converges for −1 < x < 1 but diverges for x = ±1)
n
Discuss Abel’s use of the phrase ‘x less than 1’ here (where today we would write ‘|x| < 1’).
– Abel mentions we could also have convergence for |x| ≤ 1 with lim ϕ(x) ̸= ϕ(1).
x→1
Ask students for their answers to Tasks 4 and 5 here.
IV. General concerns about foundations: If we don’t base calculus on power series, what do we use instead?
– Some possibilities (and early proponents of each):
Fluxions (Newton) ; Infinitesimals (Leibniz) ; Limits (d’Alembert) ← The “winner”!
– Chosen option of ‘limit’ raises yet another new question: What is a limit really??
14
An optional historical aside related to Discussion Item III
The use of series and power series itself was NOT new in the nineteenth century!
• Power series had been around well before the invention of calculus;
they were also part of ‘pre-calculus’ in the sense that, at least through the eighteenth century,
understanding power series was considered a pre-requisite to the study of calculus.
• Newton (and others) used power series extensively as infinite polynomials
that are easy to integrate and differentiate.
• An infinite series example from the 18th century: 1 − 1 + 1 − 1 . . . = 12
– A first “proof”:
(1 − 1) + (1 − 1) + . . . = 0 ; 1 − (1 − 1) + (1 − 1) + . . . = 1
0+1 1
Series value is the average: 2 = 2.
∞
X 1 1
(−1)n = =
1 − (−1) 2
n=1
For more about this and other divergent series in the 17th century, see the June
2006 entry of the MAA Online series How Euler Did It by Ed Sandifer (available at
http://eulerarchive.maa.org/hedi/HEDI-2006-06.pdf).
Commenting on his experience during a visit to Paris, Abel wrote the following to Holmboe on
October 24, 1826:
Legendre is an extremely amiable man, but unfortunately “as old as stones.” Cauchy
is mad and there is no way to get anywhere with him, although at present he is the
[only] mathematician who knows how to treat mathematics. His works are excellent,
but he writes in a very confused manner. In the beginning, I understood almost
nothing that he wrote, now that’s going better. . . . Cauchy is extremely Catholic
and bigoted. A very strange thing for a mathematician. . . . Poisson is a small man
with a nice little belly. He carries himself with dignity. Likewise Fourier. Lacroix is
terribly bald and remarkably old. . . . Otherwise I do not like the French as much as
the German: the French are extremely reserved with foreigners. It is very difficult to
make their close acquaintance. And I dare not count on doing so. Everyone works for
himself without caring about others. Everyone wants to teach and no one wants to
learn. The most absolute egoism reigns everywhere. The only thing the French look
for from foreigners is the practical; no one knows how to think except [the French]
themselves. The French are the only ones who can produce something theoretical.
Such are their thoughts, and you can well conceive that it is difficult to attract any
attention, especially for a beginner.
Translation prepared by the author based on original text of the letter (pp. 41–42 of Norwegian
section) and its French translation (p. 45 of French section), both given in [Holst et al., 1902].
15