Philosophy Stack Exchange is a question and answer site for those interested in the study of the fundamental nature of knowledge, reality, and existence. It only takes a minute to sign up.
Sign up to join this community
An often-quoted version of Ockham's razor (that can not be verified as being posited by Einstein himself) says "Everything should be kept as simple as possible, but no simpler."
Doesn't it violate the Ockham's razor principle? I mean the part after the comma is unnecessary and excessive, since we can not keep something more simple if it it's not possible.
The aphorism is obviously an example of surreal humor - the ascription to Einstein is somewhat shaky. I'd say that it has really nothing to do with Ockham's Razor and therefore cannot even be called a bad formulation of it.
If we grant (for the sake of argument) that Einstein is the author of the aphorism, my speculative interpretation would be the following: Einstein is not only known as a great physicist, but also as a great popularizer of science. One of the achievements of popular science is to explain scientific notions to an informed public by finding an optimal tradeoff between (1) rendering difficult notions intelligible and (2) conveying what a particular notion actually describes. In this sense one could say that the task is to simplify the notions of science without oversimplifying them - and this task (which Einstein mastered so well) is rendered, as a humorous maxim, in the aphorism quoted.
I don't think there's anything interesting here, that's just a bad formulation of Ockham's Razor (if, indeed, it is one; see DBK's answer for reasons not to think it a formulation of OR). The appropriate formulation of Ockham's Razor, and the philosophical importance of the principle (if any), is very much a matter of controversy.
See the Stanford Encyclopedia of Philosophy article on Simplicity for more information on these debates.
Simplicity is often appealed to as a theoretic virtue in both philosophy of science and metaphysics. Those who appeal to simplicity in metaphysics generally have in mind qualitative parsimony. You could formulate a version of the principle for qualitative (ontological) parsimony as follows: "don't multiply (basic or fundamental) kinds of entities beyond necessity". The force of the "beyond necessity" is to capture the ceteris paribus clause of the Razor. The simpler theory is only better if the two theories are on a par in all other important respects.
For example, David Lewis once said:
I subscribe to the general view that qualitative parsimony is good in a philosophical or empirical hypothesis; but I recognise no presumption whatever in favour of quantitative parsimony. (Lewis Counterfactuals, p. 87)
What Lewis is saying here is that fewest number of basic kinds of entities is a theoretic virtue (on his view), but that simply the fewest number of entities doesn't count in favor of a theory.
For reasons why you might want to countenance quantitative parsimony you might consult Daniel Nolan's "Quantitative Parsimony".
I think the quote is good, regardless of the actual source. It is some of a wordplay, because, as you say, simpler than possible is impossible, whereas trying to make an explanation even simpler when it is already the simplest possible is indeed possible. And that is the point, I think. One shall not sacrifice accuracy for simplicity. The best explanation is the simplest one among the most accurate ones.
My last formulations above are a bit approximate. As a Machine Learning student, I learn that model overfitting is a major concern, that is fitting a model very well to known observations, while losing performance on unseen data. ML models usually have a lot of free parameters. They are needed for representational power, and are also very useful for finding paths toward acceptable local optima. In ML research (and statistics in general), one will typically hold out subsets of the data for validation, to monitor overfitting and stop training before severe overfitting harms generalization.
The point is that sacrificing some accuracy for simplicity may be right sometimes, e.g. when the amount of data is insufficient for benefitting from hold-out validation.
Update: Here is a recommended link in response to the comment from Frank Hubeny: https://en.wikipedia.org/wiki/Overfitting
And here is an introduction to a basic validation scheme: https://en.wikipedia.org/wiki/Training,_validation,_and_test_sets
I don't think looking at this as an interpretation is the right way of looking at it: its an aphorism that gestures towards Occams Razor. Aphorisms are a literary form and are often based on some philosophical insight. They do not encapsulate the entire philosophy itself. One doesn't substitute an aphorism for the philosophy, for the entire understanding. Rather one should think of it as a comment on it.
I had understood that this aphorism had been attributed to Einstein, but given how much work is misattributed - it doesn't surprise me that there are doubts.
The part after the comma is neccessary. The aphorism would not have been memorable if it had been excised. Its saying that there is an art to choosing the simplest explanation. That perhaps the simplest explanation is not neccessarily the significant explanation. Its possibly against quantitive notions of simplicity - because there the simplest is easy to find - it is the fewest. That one has to work at finding the explanatory framework that allows for the simplest explanation. That perhaps simple explanations are actually quite complex.
Although there is a mantra about 'simple explanations' in science. This is often a matter of perspective; undoubtably a significant perspective. One for example has to look at String Theory to see a baroque explanation as also simple one. It seems to have swallowed every significant piece of mathematics over the last century (a significant piece of synthesis!), but also it expands the point of traditional science to an expanded point (ie strings, membranes etc).
Although the traditional epistemological view of this is hierarchical, and sees the baroque edifice of string theory poised on top of this expanded point - is this view not just historical and logical? Can one via Derridas deconstructionist move, invert this oppositional paradigm - for is Nature in-itself historical & logical?
The key point is where the part after comma refers to.
Better version is: Everything should be made as simple as possible, but not simpler.
Simpler refers to made and it means not to cease its coming to existence.
