You learn the flip rule. You burn it into your memory: multiply or divide by a negative number and the inequality sign flips. Then, in your next problem, you see a minus sign somewhere in the work and something panics in the back of your head — "there's a negative involved, better flip to be safe." And you flip.
That reflex is the over-correction. The flip rule is narrow. It fires only for multiplication or division by a negative. Addition and subtraction leave the inequality sign exactly where it was, no matter how many minuses appear along the way.
The rule, stated precisely
For any real numbers a, b, c:
- Addition: If a < b, then a + c < b + c. Direction preserved. Always.
- Subtraction: If a < b, then a - c < b - c. Direction preserved. Always.
- Multiplication / division by a positive c: If a < b and c > 0, then ac < bc and a/c < b/c. Direction preserved.
- Multiplication / division by a negative c: If a < b and c < 0, then ac > bc and a/c > b/c. Direction reverses — this is where the flip lives.
The flip is not triggered by the appearance of a minus sign somewhere on the page. It is triggered by one specific action: scaling both sides by a negative factor. If you did not multiply or divide by a negative, you did not trigger the flip.
Example 1: subtract a positive number — no flip
Solve x + 3 > 5.
You want x alone on the left, so you subtract 3 from both sides:
No flip. The inequality sign stayed as > throughout.
Why: you subtracted 3, a positive number. Subtraction is not multiplication by a negative. Think of it as shifting both sides leftward by the same amount — the gap between them is untouched, and the direction of the gap is untouched too.
A student who has half-learned the flip rule sometimes looks at the step "-3 on both sides" and thinks: "there's a negative — flip!" and writes x < 2. That is wrong. Test it: put x = 5 into the original. 5 + 3 = 8 > 5 is true. So x = 5 had better be in the solution set. x > 2 includes it; x < 2 excludes it. The flip was the mistake.
Example 2: subtract a negative number — still no flip
Solve x - (-5) > 2.
First simplify the left side: subtracting -5 is the same as adding 5, so the inequality is
Now subtract 5 from both sides:
No flip anywhere. Even though a negative number (-5) was sitting inside the problem, you never multiplied or divided by a negative. Subtracting, adding, and simplifying double-negatives are all flip-free operations.
Why: x - (-5) rewrites as x + 5 by the rule a - (-b) = a + b. That is an algebraic simplification of the expression on the left. It does not touch the inequality sign because you have not done the same thing to both sides — you have just renamed one side.
Example 3: where the flip actually does fire
Compare with -2x > 6. Here you divide both sides by -2:
Flip. The sign went from > to < because you divided by a negative. Notice what is different from Example 1: the negative is a multiplicative factor on x, and you undid it with division by a negative. That is exactly the operation the rule warns you about.
A decision flowchart
Why the over-correction feels tempting
Students fall into this for a reason. The phrase "when a negative is involved" is a natural but sloppy summary of the flip rule. A minus sign can show up in three very different roles:
- As a sign on a number — the constant is negative, like -5 in x - (-5) > 2. This is cosmetic. It changes values, not directions.
- As the subtraction operation — you are taking away the same amount from both sides. This preserves the inequality because the gap between the two sides is untouched.
- As a multiplicative factor — the number -c is scaling x or the whole side. This, and only this, flips the inequality when you undo it by dividing.
Only role (3) triggers the flip. The slogan "flip for negatives" collapses all three roles into one. The precise slogan is: flip if and only if you just multiplied or divided both sides by a negative number.
The geometric picture — why addition can't flip
Think of an inequality a < b as a statement about two points on the number line: a sits to the left of b. "Adding c to both sides" slides both points rightward by the same distance c (if c > 0) or leftward by the same distance |c| (if c < 0). The two points move together. Their relative positions — who is on the left, who is on the right — cannot change, because they are rigidly yoked by the shared translation.
Now think about multiplying both sides by a negative, say -1. That is not a translation. It is a reflection across zero. The point that was on the left of zero lands on the right, and vice versa. A reflection through a point genuinely swaps left-and-right — which is exactly why the inequality direction reverses. Multiplication by a negative is the only elementary operation that does this reflection; addition and subtraction are translations, which cannot.
Once you see the geometry, the rule is not a memorised boundary condition. It is the difference between sliding two points and reflecting them. Slides preserve order; reflections across a shared point reverse it.
A self-check you can run
Before flipping, ask: What did I just do to the inequality? Not what numbers are sitting in it — what operation did you perform on both sides just now? If the answer is "I added or subtracted," do not flip. If the answer is "I multiplied or divided by a positive," do not flip. Only if the answer is "I multiplied or divided by a negative" do you reverse the sign.
If you are still unsure, plug a specific value of x from your final solution into the original inequality. If it satisfies the original, your direction is correct. If it fails, you either flipped when you shouldn't have or didn't flip when you should have. This sanity check takes five seconds and catches the over-correction every time.
A worked counter-example to cement it
Solve -4 + x \ge -7 and watch how a student tempted by the over-correction goes wrong. The correct path: add 4 to both sides.
No flip. The inequality stays \ge. The solution set is [-3, \infty).
An over-corrector sees the "-4" on the left and the "-7" on the right and thinks: "there are negatives everywhere, and I'm moving a negative — flip." They write x \le -3. Check with x = 0 (which the over-corrector has excluded): -4 + 0 = -4 \ge -7 is true, so x = 0 should be in the solution. The correct answer [-3, \infty) includes 0; the flipped answer (-\infty, -3] excludes it. The over-correction costs the student every single positive value of x — half the real line or more.
The presence of minuses in the original inequality is a red herring. What matters is the operation on both sides. "Add 4" is a positive translation; it preserves the order; no flip.
The flip rule is sharp, not vague. Keep it sharp — addition and subtraction will stay flip-free, as they always were — and the arrow on your inequality will point the right way every time.