Logical Reasoning: Drawing and Testing Conclusions on the LSAT

Must Be True

What it is

A Must Be True (MBT) question asks you to find a statement that is logically forced by the stimulus. If the stimulus is true, the correct answer has to be true as well—no extra assumptions allowed.

It helps to think of the stimulus as a set of constraints. An MBT answer is something that falls out of those constraints. You are not trying to pick the most “reasonable” answer or the one that matches real life—you’re looking for what is guaranteed by what’s written.

Why it matters

MBT questions train the most fundamental LSAT Logical Reasoning skill: valid inference. Many other question types (Principle questions, some Strengthen/Weaken work, even parts of Flaw) depend on your ability to keep track of what follows from what.

Also, MBT questions punish a very common LSAT habit: treating a plausible conclusion as if it were proven. The LSAT is less interested in what sounds right and more interested in what is supported by the text.

How it works (a step-by-step method)

1. Identify the role of each sentence. On MBT, many stimuli are just a bundle of facts (premises) with no clear conclusion. Sometimes there is a conclusion, but you still treat everything given as “true for the sake of the question.”

2. Translate when needed—especially quantifiers and conditionals.

   • Quantifiers: “all,” “most,” “some,” “none,” “only,” “unless.”
   • Conditionals: “if,” “only if,” “requires,” “depends on,” “guarantees.”

3. Chain the information. If you can link statements (A implies B; B implies C), you can infer A implies C.

4. Look for safe inferences. On MBT, “safe” typically means:

   • Restatements in different words
   • Combined facts (A is true and B is true, so “A and B” is true)
   • Valid conditional/quantifier consequences

5. Attack the answer choices using denial testing. Ask: “Can I imagine a world where the stimulus is true but this answer is false?” If yes, the answer is not must-be-true.

The inference patterns MBT loves

Conditional logic (the LSAT’s favorite inference engine)

A conditional statement has a sufficient condition (the “if” part) and a necessary condition (the “then” part).

• “If A, then B” means: whenever A happens, B must happen.
• Common valid moves:
  • Contrapositive: If A → B, then not-B → not-A.
  • Chaining: If A → B and B → C, then A → C.

Common wrong moves (these often appear as tempting wrong answers):

• Affirming the consequent: A → B, B, therefore A. (Not valid.)
• Denying the antecedent: A → B, not-A, therefore not-B. (Not valid.)

Quantifiers (especially “most” vs “some”)

Quantifiers set proportions, but LSAT quantifiers are often used qualitatively:

• All: every member of a group.
• Some: at least one.
• Most: more than half.

Safe inferences:

• From all A are B, you can infer some B are A only if you know A exists. The LSAT often avoids existence guarantees unless stated.
• From most A are B, you can infer some A are B (because if more than half are, at least one is).

“Only” and “Unless”

These are classic confusion points.

• “Only if” introduces a necessary condition.
  • “A only if B” means A → B.
• “Unless” typically means “if not” or can be rewritten as an “or.”
  • “A unless B” often becomes: if not-B → A, and also can be rewritten as A or B.

Worked example 1 (MBT)

Stimulus:
All of the museum’s Renaissance paintings are kept in climate-controlled rooms. Some of the museum’s paintings are on loan to other institutions. No climate-controlled room contains any painting that is on loan.

Goal: What must be true?

Reasoning:

• All Renaissance → climate-controlled.
• Some paintings are on loan.
• No climate-controlled room contains on-loan paintings.

If Renaissance paintings are in climate-controlled rooms, and climate-controlled rooms contain no on-loan paintings, then Renaissance paintings cannot be on loan.

Must be true: No Renaissance painting is on loan to another institution.

Notice how “some paintings are on loan” is a distractor for many students—you might be tempted to connect “some” to Renaissance, but the stimulus never says that.

Worked example 2 (MBT with contrapositive)

Stimulus:
If a company’s financial statements are audited, then any major accounting error will be detected. This company’s financial statements were audited.

What must be true?

A tempting wrong thought is: “So there are no major errors.” But the conditional only says: if there is a major error, it will be detected. It does not say there are no errors.

What is forced? If the company had a major accounting error, it would be detected. So you can conclude:

Must be true: Any major accounting error the company has will be detected.

That’s basically a tight restatement of the rule applied to this company.

What goes wrong (common MBT traps)

MBT wrong answers usually fall into a few families:

• Too strong: uses “all/none/must/never” when the stimulus supports only “some/might.”
• Requires an extra assumption: sounds reasonable but isn’t forced.
• Mistaken reversal/negation: classic conditional errors.
• Scope shift: the answer talks about a different group/time/place than the stimulus supports.

Exam Focus

• Typical question patterns:
  • “Which of the following is most strongly supported by the statements above?” (Sometimes this is MBT-like; wording varies—see the MSS section for the distinction.)
  • “If the statements above are true, which of the following must also be true?”
  • A fact-set with no conclusion where you must combine constraints.
• Common mistakes:
  • Treating a plausible conclusion as guaranteed—always try to imagine a counterexample.
  • Falling for conditional reversals (“A only if B” misread as B → A).
  • Over-inferring existence (assuming a group has members when not stated).

Most Strongly Supported

What it is

A Most Strongly Supported (MSS) question asks for the answer choice that is best supported by the stimulus, even if it is not 100% logically guaranteed.

That sounds like MBT, but the mindset is slightly different:

• MBT: the correct answer is forced.
• MSS: the correct answer is the best bet—the one you should believe more than the others given the text.

In practice, MSS often still feels very “proof-driven,” just with a bit more tolerance for modest, carefully worded generalizations.

Why it matters

MSS questions test your ability to reason the way you often have to in real life: you rarely have perfect proof, but you can still draw a responsible conclusion from strong evidence.

On the LSAT, this skill also protects you from two extremes:

• Being so strict you reject the best answer because it isn’t deductively certain.
• Being so loose you pick an answer that merely “could be true.”

How it works (what to look for)

A good MSS answer typically has these features:

• It is modest in wording (often uses “some,” “likely,” “tends,” “at least one,” or a narrow claim).
• It matches the scope of the stimulus.
• It is tied to multiple pieces of evidence, not a single cherry-picked phrase.

A useful approach is:

1. Summarize the main takeaway of the stimulus in your own words.
2. Predict the type of inference you expect (causal? statistical? categorical?).
3. Compare answers by support: Which one has the strongest textual backing?
4. Eliminate “could be true” answers that aren’t particularly supported.

MSS vs MBT (a practical comparison)

Worked example 1 (MSS)

Stimulus:
In a survey of city commuters, those who began using the new express bus line reported shorter commute times on average than they reported the previous year. The express bus line did not change the number of stops on the route, but it did add a dedicated bus-only lane for part of the trip.

What is most strongly supported?

The stimulus suggests the new lane is a plausible explanation. But note what it doesn’t say: it doesn’t rule out other causes (different traffic patterns, changed work locations, memory bias in reporting, etc.). So the best supported conclusion is cautious.

Strongly supported: The dedicated bus-only lane likely contributed to the reduction in average reported commute times among those commuters.

A too-strong wrong answer would be: “The dedicated lane is the only reason commute times decreased.” The stimulus doesn’t justify “only.”

Worked example 2 (MSS with a constrained inference)

Stimulus:
All of the students who submitted the lab report on time received feedback within one week. Some students who received feedback within one week still failed the course.

What is most strongly supported?

From “some feedback-within-a-week failed,” you can safely infer:

• At least one student received feedback within one week and failed.

But you cannot infer that the student submitted on time unless you connect it—and you can’t, because receiving feedback within one week could also happen for late submissions (the first sentence doesn’t say only on-time submissions get quick feedback).

Strongly supported: Some student who received feedback within one week failed the course. (This is actually deductively forced, which can happen on MSS.)

What goes wrong (common MSS traps)

• Confusing “best supported” with “possible.” If an answer is merely consistent with the stimulus, that’s not enough.
• Choosing an answer that is emotionally/commonsense appealing but not textually grounded.
• Overcommitting to a causal story when the stimulus provides correlation or self-reports.

A helpful mental check: MSS answers tend to be the ones you’d be willing to defend in writing by quoting or paraphrasing the stimulus.

Exam Focus

• Typical question patterns:
  • “Which of the following is most strongly supported by the information above?”
  • “The statements above, if true, most strongly support which of the following?”
  • Fact patterns that invite a cautious inference rather than a strict deduction.
• Common mistakes:
  • Picking a “could be true” answer that the stimulus doesn’t actually point toward.
  • Ignoring qualifiers (“some,” “many,” “often”) and answering as if the stimulus were universal.
  • Overgeneralizing from a specific study/survey to “all people” or “always.”

Cannot Be True

What it is

A Cannot Be True (CBT) question asks you to find an answer choice that is impossible given the stimulus. If the stimulus is true, that answer choice must be false.

CBT is the mirror image of MBT:

• MBT: must be true.
• CBT: must be false.

The stimulus creates a “logic fence.” Four answer choices stay inside the fence (they could be true), and one answer choice necessarily crashes into the fence.

Why it matters

CBT questions sharpen a different but equally crucial logical skill: spotting inconsistency. In legal and analytical contexts, it’s often just as important to know what is ruled out as what is guaranteed.

CBT is also a great diagnostic for whether you truly understand conditional logic and quantifiers—many CBT correct answers exploit a single misread word like “only,” “unless,” or “most.”

How it works (a reliable strategy)

1. Treat the stimulus as fixed truth and build a compact rule set.
2. Pre-calculate the big constraints (especially contrapositives and “no/none” relationships).
3. Test each answer by trying to make it fit.
   • If you can construct a scenario where both the stimulus and the answer are true, that answer is not correct.
   • If the answer contradicts even one non-negotiable constraint, it’s the CBT.

A practical tip: CBT often rewards diagramming when the stimulus is rule-heavy (grouping-game-like logic inside LR), but you can also do it with careful verbal tracking.

The most common contradiction types

Direct contradiction

The answer states the opposite of a stimulus fact.

• Stimulus: “No A are B.”
• CBT answer: “Some A are B.”

Conditional contradiction via contrapositive

• Stimulus: A → B.
• Equivalent constraint: not-B → not-A.
• CBT answer might assert: not-B and A (which violates the contrapositive).

Quantifier mismatch

• Stimulus supports “some,” answer claims “all,” and that “all” claim cannot be made true without violating another constraint.

Worked example 1 (CBT)

Stimulus:
Every employee who has access to the secure server has completed cybersecurity training. No contractor has completed cybersecurity training.

What cannot be true?

Let’s translate:

• Access → Training.
• Contractors → not-Training.

From these, you can infer: Contractors → not-Access (contrapositive of Access → Training combined with contractors’ lack of training).

So the impossible situation is:

• A contractor has access to the secure server.

That contradicts the derived rule that no contractor can have access.

Worked example 2 (CBT with “unless”)

Stimulus:
A package will be delivered tomorrow unless the weather service issues a severe storm warning.

Rewrite: If no severe storm warning, then delivered tomorrow. Also equivalent: Delivered tomorrow or severe storm warning.

What cannot be true?

• “No severe storm warning is issued, and the package is not delivered tomorrow.”

That violates the conditional: not-warning → delivered.

What goes wrong (common CBT traps)

• Stopping at “not supported.” In CBT you need ruled out, not merely “not proven.”
• Forgetting to use contrapositives. Many impossibilities are hidden one step away.
• Confusing “could be false” with “must be false.” You’re looking for the one answer that cannot be made consistent.

Exam Focus

• Typical question patterns:
  • “Which of the following cannot be true?”
  • “If the statements above are true, which of the following must be false?”
  • Rule-heavy stimuli where one answer violates a key constraint.
• Common mistakes:
  • Eliminating an answer just because it isn’t directly mentioned (irrelevant to CBT).
  • Missing the contrapositive and therefore missing the contradiction.
  • Misreading “unless/only if” and building the wrong rule set.

Resolve the Paradox

What it is

A Resolve the Paradox question gives you a set of statements that seem inconsistent—together they create a puzzle. Your job is to choose an answer that, if true, makes the situation make sense.

Crucially, you are not “proving” the answer from the stimulus. Instead, you are looking for information that reconciles the facts by showing how they can coexist.

Why it matters

Resolution questions test flexible reasoning. In real arguments (and in law), you often face two credible facts that point in opposite directions. The skill is not to deny one fact, but to find a missing distinction—different groups, different time frames, different definitions, or an overlooked causal factor.

These questions also build discipline about what the correct answer must do: it must address the specific tension in the stimulus, not provide random additional facts.

How it works (the resolution mindset)

1. State the paradox clearly. Put it in your own words as: “How can both of these be true?”

2. Identify what would count as a resolution. Typically, the right answer will:

   • Add a distinction (different groups, times, conditions)
   • Reveal a hidden factor (a third variable)
   • Clarify an ambiguous term (two meanings of the same word)
   • Show a compensating effect (one change offsets another)

3. Predict the shape of the fix before reading choices deeply. For example: “Maybe the measurement changed,” or “Maybe the people counted in group A aren’t the same as group B.”

4. Choose the answer that directly dissolves the conflict with minimal extra storytelling. The best resolutions tend to be efficient: they patch the exact leak and don’t introduce new mysteries.

Common resolution “templates” (high-yield patterns)

Different groups (selection effects)

The stimulus appears contradictory because you assume the same population is involved in both statements.

• Fix: show the statements refer to different subsets.

Different times

A trend looks inconsistent because one fact is from earlier, the other from later.

• Fix: show a timing change (policy implemented mid-period, delayed effects).

Different definitions / shifting standards

One term is used in two senses.

• Fix: clarify definition or measurement criteria.

A third factor (confounding variable)

Two variables seem linked, but a hidden cause changes the picture.

• Fix: introduce the confounder.

Net effect (two forces in opposite directions)

One change should increase something, but another simultaneous change decreases it more.

• Fix: identify the stronger countervailing influence.

Worked example 1 (Resolve the Paradox)

Stimulus:
A restaurant introduced a new menu designed to reduce wait times by offering fewer made-to-order items. However, average wait times increased after the new menu was introduced.

Paradox: A change intended to reduce wait times was followed by longer wait times.

A resolving answer might say:

• The new menu attracted significantly more customers, increasing demand beyond the kitchen’s capacity.

That makes both facts compatible: fewer made-to-order items can speed service per order, but higher volume can still increase average waits.

Notice what would not resolve it:

• “The new menu has healthier options.” (Irrelevant to the conflict.)
• “Some customers like the old menu better.” (Doesn’t explain longer waits.)

Worked example 2 (Resolve with definition shift)

Stimulus:
A company reports that workplace injuries fell by 20% last year. Yet employee surveys show that more employees than ever say their jobs have become physically dangerous.

Paradox: Fewer reported injuries, but more perceived danger.

A strong resolution:

• The company changed its reporting policy so that minor injuries are no longer classified as “workplace injuries” in its official statistics.

Now both can be true: official injury numbers fall due to classification changes, while perceived danger rises.

What goes wrong (common resolution traps)

• Picking an answer that strengthens one side instead of reconciling both. If it makes injuries even lower without explaining the increased danger perception, it’s not resolving.
• Choosing an answer that contradicts the stimulus. The right answer should allow all stimulus statements to remain true.
• Overcomplicating. Students sometimes prefer a dramatic story, but LSAT resolutions are usually simple: one missing piece that bridges the gap.

A helpful check: After you pick an answer, you should be able to say in one or two sentences, “Because of this, statement A and statement B can both be true.” If you can’t produce that bridge, reconsider.

Exam Focus

• Typical question patterns:
  • “Which of the following, if true, most helps to resolve the apparent discrepancy?”
  • “Which of the following explains how the statements above could both be true?”
  • Two facts that look inconsistent, often involving trends, surveys, or cause-and-effect expectations.
• Common mistakes:
  • Selecting a choice that is merely related to the topic but doesn’t address the contradiction.
  • Treating it like Strengthen/Weaken and arguing for one side rather than reconciling both.
  • Missing a key word that indicates the real tension (e.g., “average,” “reported,” “last year,” “some”).