Dear friends, Together with many others, I am deeply saddened by the death of George Mackey. For far too long, I had been delaying the trip to see him and continue our discussions which had started in the "Weyl'sche Kammer" at the ETH in Zurich and continued in the physics center in Trieste. Long before I met him, his insights into mathematics and into quantum mechanics had been informing my own thinking. It was the study of his book on quantum mechanics in 1967 which led directly to the joint course by Saunders Mac Lane and me at the University of Chicago. But his relation to category theory goes back much further than that, as Saunders and Sammy had explained to me. George Mackey's Ph.D. thesis displayed remarkable thinking of a categorical nature, even before categories had been defined. Specifically, the fact that the category of Banach spaces and continuous linear maps is fully embedded into a category of pairings of abstract vector spaces, together with the definition and use of "Mackey convergence" of a sequence in a "bornological" vector space were discovered there and have played a basic role in some form in nearly every book on functional analysis since. What is perhaps unfortunately not clarified in nearly every book on functional analysis, is that these concepts are intensely categorical in character and that further enlightenment would result if they were so clarified. And who, despite initial skepticism, permitted the first paper giving an exposition of the theory of categories to see the light of day in the Transactions of the AMS in 1945? None other than the referee, George Whitelaw Mackey. Sincerely, F. William Lawvere ************************************************************ F. William Lawvere Mathematics Department, State University of New York 244 Mathematics Building, Buffalo, N.Y. 14260-2900 USA Tel. 716-645-6284 HOMEPAGE: http://www.acsu.buffalo.edu/~wlawvere ************************************************************
I had not heard that Mackey had died and am also saddened, although I never met him. Bill does not mention that Mackey's thesis, I think it was published in the same volume of the Transactions as "General Theory of natural equivalances", was the direct source of the Chu construction. As Bill mentioned the category of pairs embeds the category of Banach spaces (and continuous linear maps). It is in fact equivalent to the category of what are now called Mackey spaces, which are characterized as the locally convex topological vector spaces that have the finest possible topology for their set of continuous linear functionals. I once wrote to Mackey asking him if he had had any intention of considering a space of linear functionals as a _replacement_ for a topology or merely an adjunct to it. Unfortunately, he did not reply, even to a snail mail "letter" (as we used to call them). As it happens it was just yesterday that I sent off the followin abstract of the talk I will give in the category session of the Canadian Math. Soc. meeting in Calgary in early June: A standard theorem says that any locally convex topological vector space has a finer topology, its \emph{Mackey topology} with the same set of continuous linear functionals and that is the finest possible topology with that property. If $E$ and $F$ are two such spaces topologize the space $\mbox{Hom(E,F)}$ of continuous linear transformations $E\to F$ with the weak topology induced by the algebraic tensor product $E\otimes F'$ and then let $[E,F]$ denote the associated Mackey topology. Let $F^*$ denote the dual $F'$ topologized by the Mackey topology on the weak dual and let $E\otimes F=[E,F^*]^*$ (whose underlying vector space is the algebraic tensor product). Then for any Mackey spaces $E$, $F$, and $G$, \begin{enumerate} \item $[E\otimes F,G]\cong [E,[F,G]]$ \item $E\cong E^*{}^*$ \item $[E,F]\cong (E\otimes F^*)^*$ \end{enumerate} which is summarized by saying that the category of Mackey spaces and continuous linear transformations is $*$-autonomous. This category is equivalent to the category of weakly topologized locally convex topological vector spaces (which have the coarsest possible topology for their set of continuous linear functionals) which is therefore also $*$-autonomous. They are also equivalent to the chu category of vector spaces (which will be explained). %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% On Thu, 23 Mar 2006, F W Lawvere wrote:
Dear friends,
Together with many others, I am deeply saddened by the death of George Mackey. For far too long, I had been delaying the trip to see him and continue our discussions which had started in the "Weyl'sche Kammer" at the ETH in Zurich and continued in the physics center in Trieste. Long before I met him, his insights into mathematics and into quantum mechanics had been informing my own thinking. It was the study of his book on quantum mechanics in 1967 which led directly to the joint course by Saunders Mac Lane and me at the University of Chicago. But his relation to category theory goes back much further than that, as Saunders and Sammy had explained to me.
George Mackey's Ph.D. thesis displayed remarkable thinking of a categorical nature, even before categories had been defined. Specifically, the fact that the category of Banach spaces and continuous linear maps is fully embedded into a category of pairings of abstract vector spaces, together with the definition and use of "Mackey convergence" of a sequence in a "bornological" vector space were discovered there and have played a basic role in some form in nearly every book on functional analysis since. What is perhaps unfortunately not clarified in nearly every book on functional analysis, is that these concepts are intensely categorical in character and that further enlightenment would result if they were so clarified.
And who, despite initial skepticism, permitted the first paper giving an exposition of the theory of categories to see the light of day in the Transactions of the AMS in 1945? None other than the referee, George Whitelaw Mackey.
Sincerely, F. William Lawvere
************************************************************ F. William Lawvere Mathematics Department, State University of New York 244 Mathematics Building, Buffalo, N.Y. 14260-2900 USA Tel. 716-645-6284 HOMEPAGE: http://www.acsu.buffalo.edu/~wlawvere ************************************************************
Dear Mike, Looking into the MathSciNet, I see that: Mackey's 1942 thesis was published in abbreviated form in the Proceedings of the National Academy of Sciences, vol. 29 (1943). The Math Reviews reviewer seems to clearly understand that part of idea was to replace open sets with linear functionals. The extended publication in the Transactions was in vol. 57 (1945) whereas the publication of Eilenberg and Mac Lane's famous paper was in vol. 58. In the same volume 58 there is a paper by Clifford Truesdell. Best, Bill ************************************************************ F. William Lawvere Mathematics Department, State University of New York 244 Mathematics Building, Buffalo, N.Y. 14260-2900 USA Tel. 716-645-6284 HOMEPAGE: http://www.acsu.buffalo.edu/~wlawvere ************************************************************ On Fri, 24 Mar 2006, Michael Barr wrote:
I had not heard that Mackey had died and am also saddened, although I never met him.
Bill does not mention that Mackey's thesis, I think it was published in the same volume of the Transactions as "General Theory of natural equivalances", was the direct source of the Chu construction. As Bill mentioned the category of pairs embeds the category of Banach spaces (and continuous linear maps). It is in fact equivalent to the category of what are now called Mackey spaces, which are characterized as the locally convex topological vector spaces that have the finest possible topology for their set of continuous linear functionals.
I once wrote to Mackey asking him if he had had any intention of considering a space of linear functionals as a _replacement_ for a topology or merely an adjunct to it. Unfortunately, he did not reply, even to a snail mail "letter" (as we used to call them).
As it happens it was just yesterday that I sent off the followin abstract of the talk I will give in the category session of the Canadian Math. Soc. meeting in Calgary in early June:
A standard theorem says that any locally convex topological vector space has a finer topology, its \emph{Mackey topology} with the same set of continuous linear functionals and that is the finest possible topology with that property. If $E$ and $F$ are two such spaces topologize the space $\mbox{Hom(E,F)}$ of continuous linear transformations $E\to F$ with the weak topology induced by the algebraic tensor product $E\otimes F'$ and then let $[E,F]$ denote the associated Mackey topology. Let $F^*$ denote the dual $F'$ topologized by the Mackey topology on the weak dual and let $E\otimes F=[E,F^*]^*$ (whose underlying vector space is the algebraic tensor product). Then for any Mackey spaces $E$, $F$, and $G$, \begin{enumerate} \item $[E\otimes F,G]\cong [E,[F,G]]$ \item $E\cong E^*{}^*$ \item $[E,F]\cong (E\otimes F^*)^*$ \end{enumerate} which is summarized by saying that the category of Mackey spaces and continuous linear transformations is $*$-autonomous.
This category is equivalent to the category of weakly topologized locally convex topological vector spaces (which have the coarsest possible topology for their set of continuous linear functionals) which is therefore also $*$-autonomous. They are also equivalent to the chu category of vector spaces (which will be explained).
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
On Thu, 23 Mar 2006, F W Lawvere wrote:
Dear friends,
Together with many others, I am deeply saddened by the death of George Mackey. For far too long, I had been delaying the trip to see him and continue our discussions which had started in the "Weyl'sche Kammer" at the ETH in Zurich and continued in the physics center in Trieste. Long before I met him, his insights into mathematics and into quantum mechanics had been informing my own thinking. It was the study of his book on quantum mechanics in 1967 which led directly to the joint course by Saunders Mac Lane and me at the University of Chicago. But his relation to category theory goes back much further than that, as Saunders and Sammy had explained to me.
George Mackey's Ph.D. thesis displayed remarkable thinking of a categorical nature, even before categories had been defined. Specifically, the fact that the category of Banach spaces and continuous linear maps is fully embedded into a category of pairings of abstract vector spaces, together with the definition and use of "Mackey convergence" of a sequence in a "bornological" vector space were discovered there and have played a basic role in some form in nearly every book on functional analysis since. What is perhaps unfortunately not clarified in nearly every book on functional analysis, is that these concepts are intensely categorical in character and that further enlightenment would result if they were so clarified.
And who, despite initial skepticism, permitted the first paper giving an exposition of the theory of categories to see the light of day in the Transactions of the AMS in 1945? None other than the referee, George Whitelaw Mackey.
Sincerely, F. William Lawvere
************************************************************ F. William Lawvere Mathematics Department, State University of New York 244 Mathematics Building, Buffalo, N.Y. 14260-2900 USA Tel. 716-645-6284 HOMEPAGE: http://www.acsu.buffalo.edu/~wlawvere ************************************************************
I met George Mackey in April 1967 at the British Mathematical Colloquium in Swansea, where I gave an invited talk on the groupoid van Kampen theorem, and I overheard some people in the common room saying they were not completely convinced. You win some, you lose some! But Mackey came up to me at tea and said: `That was very interesting. I have been using groupoids for years. My name is Mackey.' He then told me of his work in ergodic theory using `virtual groups'. It occurred to me that if the groupoid idea can be arrived at from two quite different directions, then there couild be more in the groupoid idea than met the eye. It became clear that he used the action groupoid of a group action, and this convinced me that I should add to my planned book a chapter on covering spaces and covering groupoids. For those who are unaware of the idea of a virtual group, Mackey's idea was that since a transitive action of a group corresponded to a conjugacy class of subgroups, then an ergodic action (i.e. one where the orbits are of measure 0 or 1) should correspond to an analogous concept. His exposition went through various phases, including a cocycle formulation, and eventually involved the measured groupoid corresponding to an ergodic action. This work, and that of his students, such as Arlan Ramsay, has been a foundation, as I understand it, for much work on the C^*-algebras of measured groupoids. We met a few more times, and he was always most friendly and genuine. When I went to Bangor, Tony Seda came there from Warwick in order to help his mother who was ill and lived in Llandudno. His MSc project had been in measure theory. So we agreed he should look at Mackey's work. In the end he developed Haar measure in this area, and when I told Mackey he said *his* student was doing the same! Tony's excellent papers in this area have perhaps not been as well noticed as they should, so Tony in the end moved into theoretical computer science. So my conclusion is that George Mackey was a great pioneer in structural ideas in mathematics, with a broad range of interests, and a really nice guy. Ronnie Brown
participants (3)
-
F W Lawvere -
Michael Barr -
Ronnie Brown