the order of operations is arbitrary.
Not at all. Algebra was invented by the Arabs long ago, but it did not come into general use in Europe till the 17th century. Of course the symbols used were somewhat arbitrary, but not the logical order. Here's an excerpt from Frans van Schooten's 1646 edition of Vieta:
________________
B in D quad. + B in D
We would express this as B(D^2 + BD). Van Schooten uses a horizontal bar (viniculum) over a sum whereas we use parenthesis. (Canadians use brackets.) All of these symbols are equally arbitrary, but the logic behind them is not. What is important here? The fact that Van Schooten even felt the need to use the vertical bar suggests that even as long ago as 1646 the idea that Multiplication/division was of a higher order than addition/subtraction!
I'm no historian, but I do like history and have actually looked into this subject a number of years ago. It seems to me that the basic rule that multiplication (and division) has precedence over addition (and subtraction) arises naturally and without much disagreement in the 17th-century literature as algebraic notation was being developed and the need for such conventions arose. Even though there were numerous competing systems of symbols, forcing each author to state his conventions at the start of a new book, they did not have much to say on the subject of order of operations. This is because the distributive property implies a natural hierarchy in which multiplication is more powerful than addition.
(Note the absence of a separate vinculum upon "B in D" above. This seems to indicate already that it is assumed by Van Schooten that the reader would understand that multiplication happens at a higher order than addition.)
Also, there may be something in the grammar of Arabic that logically helped with the development of algebraic formalism. I do not read or understand Arabic well enough to parse that. (My studies are limited to the "Teach Yourself Arabic" paperback with cassette tape I purchased in anticipation of a trip to the Pyramids, and that was a very long time ago and I have forgotten most of what I learned then.) Or it may be that the language of mathematics, independent of symbolism, transcends the spoken idiom, and that all learned men in all cultures would naturally assign a higher order to multiplication than to addition.
Ernest knows a whole bunch of stuff about a whole bunch of stuff. Maybe he knows whether the we think multiplication trumps addition because the Arabs taught us that, or whether the Europeans and Chinese figured that out independently because it just feels so right.