This article was inspired by a response to a reader’s comment on one of my other PEMDAS/OOO articles. I thought the point I made was worthy enough to be a separate post.
The idea that a fraction is identical to a division problem with an obelus just doesn’t hold water if you analyze the uses of fractions. For example, if I say I ate two-thirds of an apple pie, I’m not intending that to be a division problem. It’s a ratio indicating how much of the pie I ate, i.e., it’s a stand-alone value, or a monomial. We must consider as well that a fraction is more accurate than a decimal if the decimal value has a repeating pattern. Perhaps you’re familiar with the proof that 0.999999…. (repeating to infinity) equals 1.0. If 1/3 (0.333…..) + 2/3 (0.666…..) = 1.0, then the sum of the decimal equivalents must equal 1. Via mathematical proof, this is proven true. However, you never actually get to the point of being able to “carry the 1,” so the decimal equivalent of the sum of the fractions is 10^(-∞) off.
Granted, such a difference in real measurement is not humanly perceptible, but it does reveal a potential issue when talking about astronomical or microscopical measurements, for example. Even if the fraction is part of a larger expression or formula, it’s always possible that the denominator of the fraction will “cancel out” exactly as a common factor with another element in the expression, so the simplification process is the “division,” but it may not be with the numerator of the fraction. For example, if I have the expression 6 x (2/3), I “cancel” the common factor of three from the 6 (leaving 2) and the denominator 3 (leaving 1, thus a whole number), so I’m left with 2 x 2 = 4. Division is happening in the cancellation of the common factor, but it happens with an element outside of the fraction itself. Therefore, the fraction itself is NOT a division problem. There are other elements acting on the fraction.
What is key to the monomial discussion, then, is the juxtapositional binding. In the expression 8 ÷ 2(2 + 2), for the sixteeners to claim that somehow the 8 ÷ 2 becomes a fractional coefficient is contrary to the concept of the fraction. The obelus doesn’t have the binding property inherent with the fraction construction, so to replace the obelus with a fraction bar and say “they’re grouped” while unbinding the 2 from the (2 + 2) defies the clearly indicated juxtapositional binding. They try to have it both ways, but for whatever reason, they’re not seeing that. (Personally, I think they got burned at some point thinking they were justified in adding the multiplication sign and they got corrected by someone who believes as we do, and instead of thinking it through, they hunkered down in their rebellion, but that’s another story.)
This is where the linguistic aspect comes into play. We’ve seen how Wolfram interprets “Eight divided by twice the sum of two plus two” as 1, but “Eight divided by two times the sum of two plus two” as 16. “Twice” and “two times” mean the same thing at face value, but “twice” binds to the “two plus two” as an adverbial modifier, while “two times” is treated as a subject-verb combination. The latter treats “two” as the subject and “times” as the verb substituting for the multiplication sign, thus 2 x (2 + 2). “Twice” has the multiplication implied in the adverbial phrase “twice the sum of two plus two,” so this more accurately reflects the implied multiplication of 2(2 + 2). As such, 2(2 + 2) is the mathematical equivalent of an adverbial phrase intended to be taken as a semantic unit, with or without the external parentheses, just like a fraction by itself is not a division problem, but a singular or monomial value. Therefore, 8 ÷ 2(2 + 2) = 1. Period. End of debate.
In today’s educational environment, there is a lack of critical thinking development. Young people are trained in mathematics simply to get to the point of an answer, not to learn the theory behind mathematics. In other words, they’re trained like calculators, and in this instance, they’re trained to process a problem like a cheap calculator would solve it instead of taking a broader view of the linguistic aspects of any mathematical expression. It’s really no different than giving a non-English speaker an English text and English dictionary and asking them to translate without any concept of grammatical rules, syntax, idioms, etc. PEMDAS/OOO is a formula designed to follow how a calculator solves an expression, not how students of math have solved the expressions in the past and certainly not to make the paradigm consistent with algebra, which treats an expression like 8 ÷ 2a as 8 ÷ (2a) and NOT 8 ÷ 2 x a. The linguistic argument supports the algebraic view of such expressions and therefore supports the same view for arithmetic.
Scott Stocking
My views are my own.
Sunday Morning Greek Blog 