Search on the web for “as simple as possible, but no simpler” and the web will tell you that Albert Einstein said this. Einstein said a lot of things, and a lot of things are attributed to him; knowing whether he, in fact, said a particular pithy quote is a bit of a problem.
I spent a good couple of hours this morning tracking this down, and I think that the Wikiquote page on Einstein gets it right. He did write:
It can scarcely be denied that the supreme goal of all theory is to make the irreducible basic elements as simple and as few as possible without having to surrender the adequate representation of a single datum of experience.
Unfortunately, the original source, although online, is behind JSTOR’s paywall, although I was able to read it because I was searching online while at Kalamazoo College’s Upjohn Library. In the Wikquote discussion page on Einstein, another similar quotation appears, but in slightly different variants. “JeffQ” asked if someone would look up the exact quotation in the “Autobiographical Notes” Einstein wrote for a 70th birthday Festschrift called Albert Einstein: Philosopher-Scientist. I trucked up to the third floor and found the quotation. The original is in German:
Eine Theorie ist desto eindrucksvoller, je grösser die Einfachheit ihrer Prämissen ist, je verschiedenartigere Dinge sie verknüpft, und je weiter ihr Anwendungsbereich ist.
with the translation (by the book’s editor, Paul Arthur Schilpp) on the facing page:
A theory is the more impressive the greater the simplicity of its premises is, the more different kinds of things it relates, and the more extended is its area of applicability.
Wikiquotes notes that this is similar to Occam’s Razor:
Entia non sunt multiplicanda praeter necessitatem (Entities are not to be multiplied beyond necessity).
We may assume the superiority ceteris paribu of the demonstration which derives from fewer postulates or hypotheses
and it gives similar quotes from Aquinas, Kant, Newton, Galileo, Lavoisier, and (of course), Einstein (not the faux quotation, but another altogether).
The article’s rewrite of Entia non sunt muliplicanda praeter necessitatem is the delightfully logistic:
Other things being equal, if T1 is more ontologically parsimonious than T2 then it is rational to prefer T1 to T2.
So, what is the simplest way to express Occam’s Razor? I think “as simple as possible, but no simpler,” which we must attribute to the wisdom of the crowd rather than to any one person, is about as good as one can get.