4
$\begingroup$

Frequently, 19th century physicists—e.g., Helmholtz or Maxwell—did not use modern-day vector notation, which Gibbs contributed in large part to.

For example, Helmholtz in his famous paper on the conservation of energy writes

$$X=m\frac{du}{dt},\hspace{1em}Y=m\frac{dv}{dt},\hspace{1em}Z=m\frac{dw}{dt},$$

where $X,Y,Z$ are components of a force and $u=dx/dt$, $v=dy/dt$, $w=dz/dt$ are the components of the tangential velocity $q$. Instead, he could've written this much more concisely as:

$$\vec{F}=m\vec{a},$$

where $\vec{F}=(X,Y,Z)$ and $\vec{a}=(u,v,w)$.

Similarly, he could've written

$$d(q^2)=\frac{d(q^2)}{dx}dx+\frac{d(q^2)}{dy}dy+\frac{d(q^2)}{dz}dz$$

much more concisely as

$$d(q^2)=\nabla q^2\cdot d\vec{r}.$$

Thus:

  1. Why did those physicists use such a cumbersome, redundant way of writing what today we'd write with vector notation?
  2. Are there some expressions that cannot be expressed in vector notation but must be expressed in this seemingly cumbersome notation that predated vector notation?

In summary:

  1. What were the criticisms against the introduction of "vector analysis"?
$\endgroup$
  • 4
    $\begingroup$ Wrong. It's a sign of progress that eventually people realized that vectors could be condensed into a single letter without losing their essential features, i.e., suppressing coordinates can be a good idea. I don't agree that the first people to work with vectors should realize this. $\endgroup$ – KCd Dec 2 '15 at 8:00
  • 3
    $\begingroup$ Strange question. Why did not they (19s century mathematicians) travel by airplanes which is much faster than ships and rail? $\endgroup$ – Alexandre Eremenko Dec 2 '15 at 14:06
  • 2
    $\begingroup$ Did you mean why didn't they use even after it became available? Maxwell actually did use Hamilton's quaternionic notation, a precursor of vector notation, in some of his writings, but Gibbs and Heaviside developed vector analysis mostly after Helmholtz and Maxwell were scientifically active. $\endgroup$ – Conifold Dec 2 '15 at 19:36
  • 1
    $\begingroup$ I don't see anything terribly wrong about the question itself except that it reflects a rather puzzlingly blatant misunderstanding on the part of the OP. And that's what we're here to fix. +1 $\endgroup$ – silvascientist Dec 3 '15 at 8:56
  • 2
    $\begingroup$ I'm saying "wrong" in response to your suggestion that the 19th century physicists could have used "more concise notation, right?" The tone of the question sure sounds as if you don't understand why those physicists did not use a notation that was not even around at that time. They expressed things very computationally (coordinates) because nobody realized that suppressing computational information was actually a good idea. You could ask in the same way why Fermat, Euler, and others did not create the notation of modular arithmetic developed by Gauss 100+ years later. $\endgroup$ – KCd Dec 3 '15 at 23:13
4
$\begingroup$

You can see :

Josiah Willard Gibbs' Elements was privately printed in 1881 and 1884.

His pupil Edwin Bidwell Wilson compiled the textbook Vector Analysis, based on Gibbs' lectures, as Gibbs was at the time busy preparing his book on thermodynamics :

Vector Analysis, first published in 1901 and based on the lectures that Josiah Willard Gibbs had delivered on the subject at Yale University, did much to standardize the notation and vocabulary of three-dimensional linear algebra and vector calculus, as used by physicists and mathematicians. It went through seven editions (1913, 1916, 1922, 1925, 1929, 1931, and 1943).

Oliver Heaviside was the other co-founder of vector analysis :

Heaviside did much to develop and advocate vector methods and the vector calculus. Maxwell's formulation of electromagnetism consisted of 20 equations in 20 variables. Heaviside employed the curl and divergence operators of the vector calculus to reformulate 12 of these 20 equations into four equations in four variables, the form by which they have been known ever since.

See again Crowe's book : Ch.5.VII. Heaviside's Electrical Papers, page 163-on :

The first paper in which Heaviside introduced vector methods was his 1882-1883 paper "The Relations between Magnetic Force and Electric Current," published in the Electrician. The way in which vectors were introduced by Heaviside is somewhat surprising.


J.C.Maxwell's A Treatise on Electricity and Magnetism was published in 1873, prior to Gibbs's pamphlet.

The same for Hermann von Helmholtz :

Helmholtz studied the phenomena of electrical oscillations from 1869 to 1871, and in a lecture delivered to the Naturhistorisch-medizinischen Vereins zu Heidelberg (Natural History and Medical Association of Heidelberg) on April 30, 1869 titled On Electrical Oscillations [...]. In 1871, Helmholtz moved from Heidelberg to Berlin to become a professor in physics. He became interested in electromagnetism and the Helmholtz equation is named for him.

Regarding Maxwell's attitude, see Crowe : Ch.5.V. James Clerk Maxwell: Critic of Quaternions, page 127-on :

Of the four books written by Maxwell after 1873 only one mentioned vectors; this is his elementary work on mechanics published in 1876 and entitled Matter and Motion. Herein Maxwell included a short section on the idea of and the addition and subtraction of vectors. Quaternions were not mentioned [page 138].

$\endgroup$
  • $\begingroup$ grazie per la citazione a Crowe $\endgroup$ – Geremia Dec 4 '15 at 5:02
4
$\begingroup$

Pierre Duhem (1861-1916) gave in his The Aim & Structure of Physical Theory (1906) p. 77—commenting on what he considered a characteristic of, generally, English thinkers, i.e., their "ampleness of mind" (Pascal's esprit de géométrie, vs. the esprit de finesse characteristic of a "French [or German] mind")—a criticism of "vector analysis":

But only among the English is ampleness of mind found so frequently as an endemic, traditional habit; thus it is only among English men of science that symbolic algebras, the calculus of quaternions, and "vector analysis" are customary, most of the English treatises making use of these complex and shorthand languages. French and German mathematicians do not learn these languages readily; they never succeed in speaking them fluently or, above all, in thinking directly in the forms which constitute these languages. In order to follow a calculation based on the method of quaternions or of "vector analysis" they have to translate it into a version of classical algebra. One of the French mathematicians, Paul Morin, who had studied most profoundly the different kinds of symbolic calculi, once told me: "I am never sure of a result obtained by the method of quaternions until I have checked it by our old Cartesian algebra."

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.