TOEIC Link Reading — Sole Discretion And Reasonable Discretion Decoding Under The Discretion Notice: How To Read A Power That Must Be Exercised Reasonably As Constrained Rather Than As Unlimited, And Stop Reading Every Grant Of Discretion As A Free Hand

The TOEIC Link reading section builds a recurring passage type around discretion clauses — the sole discretion that lets a party decide without giving reasons, the reasonable discretion that binds the decision to an objective standard, the good-faith duty that constrains even a broadly worded grant, the discretion that reads as absolute on its face but carries a qualifier that limits it — and constructs its high-discrimination questions around whether the discretion is constrained rather than around whether a power to decide was granted. The band-ceiling candidate reads a clause granting a party discretion, sees a power to choose and no visible list of conditions, and concludes the party may act however it likes, because a grant of discretion reads like a free hand. The candidate is scored wrong because a discretion qualified by "reasonable," by a good-faith duty, or by a stated standard is a bounded power: the party may decide, but the decision must meet the limit the clause installs, and the reader who treats every grant as unlimited has skipped the qualifier that constrains it. The power to decide is real, but the scope of the power is fixed by whether a standard bounds it.

The scoring consequence is that the bare grant of discretion functions as an attractor. The passage presents a clause giving a party discretion, states the power plainly, the question asks whether the party may make a particular decision, and the answer choice that reads the discretion as unlimited is offered as the trap. The candidate who reads any grant of discretion as absolute rewards the breadth of the word and selects the trap; the candidate who reads for the constraint checks whether a reasonableness standard, a good-faith duty, or a stated qualifier bounds the power and, finding the discretion limited, treats the decision as valid only if it meets the standard, selecting the answer the clause actually supports. This guide formalizes the constrained-versus-unlimited reading model that reframes a grant of discretion from a free hand into a power subject to a limit, the discretion-decoding protocol for clauses that grant a power under a standard, and the four-week drill that installs the discipline of testing the constraint on a discretion before treating any grant as absolute.

Why a grant of discretion reads as unlimited and functions as bounded

The discretion notice presents a surface that invites the unlimited reading. It grants the power to decide, it states the grant with the breadth of an open authority, and it places the operative constraint — the reasonableness standard, the good-faith duty, the qualifier that the decision must be justified — in wording that reads like a description of how the power is exercised rather than a limit on whether the exercise is valid. The candidate who reads the notice for the grant forms the impression that the party may act freely, and then answers the decision question as though the bare power decided the scope. The word is the wrong anchor. The notice is not recording an unlimited authority; it is recording a power to decide that a standard bounds, and whether a particular decision is valid turns on whether it meets the limit, not on whether a discretion was granted.

The gap between the unlimited reading and the constrained reading is where the discrimination lives. A party may hold discretion and still be bound to exercise it reasonably or in good faith, so that a decision outside the standard is not a valid exercise of the power at all; the grant is real but never free. The question is constructed to describe exactly this limit: a discretion granted, a standard that qualifies it, competing with a reader's expectation that a power to decide is a power to decide anything, so that the candidate who fixes on the grant treats a bounded discretion as an absolute one. The candidate who reads any grant as unlimited rewards the clause for the breadth of its language; the candidate who reads for the constraint traces the standard the qualifier installs and treats only a conforming decision as valid. For the related discipline of reading a duty to try hard rather than a duty to succeed, see the reading best efforts and reasonable efforts decoding under the performance-standard notice guide, and for reading a right that may be given up by conduct, see the reading waiver and non-waiver decoding under the preserved-right notice guide.

The reframe from grant-as-free to grant-as-bounded is the central correction. The discretion notice is a specification of a power subject to a limit — the authority to decide, the standard the decision must meet, the good-faith duty that constrains even a broad grant — and no decision is valid merely because a discretion was granted. The candidate must read every discretion clause and test whether a standard bounds it before treating the power as unlimited. The reframe is installable, and the discretion-decoding protocol below operationalizes it for the common case where the constraint, not the grant, is the fact that decides.

The discretion-decoding protocol

The discretion that genuinely runs free — the sole discretion expressly freed from any reasonableness duty, the decision the clause says need not be justified — is common enough to be plausible, but the test constructs its hardest items around discretions granted in broad words yet bounded by a standard, because that gap between an open grant and a constrained power is where the discrimination the protocol exists to navigate is built. The discretion-decoding protocol has three steps.

The first step is to locate the grant and read whether a standard qualifies it. The candidate reads past the word "discretion" and identifies whether a reasonableness qualifier, a good-faith duty, or a stated standard bounds the power, or whether the discretion is expressly freed from any limit. The most common extraction failure is registering the grant while skimming past the qualifier that constrains it, which converts a bounded power into a free hand by default. The standard must be read because the question will turn on whether the decision meets it, not on whether a discretion was granted.

The second step is to test the decision against the standard. The candidate isolates the particular decision the question raises and asks whether it meets the reasonableness or good-faith limit the clause installs, setting aside the breadth the word "discretion" projects. The most common outcome failure is letting the bare grant stand in for validity, when a standard bounds the power and only a conforming decision is a valid exercise of it. The limit must be tested because the question will turn on the standard, not on the grant.

The third step is to read the answer for the constraint rather than the grant. The candidate selects the choice that reflects a discretion bounded by its standard, not the choice that treats the power as unlimited. The most common selection failure is choosing the answer that rewards the breadth of the grant, when the clause has qualified the discretion and the question asks whether a particular decision is valid. The answer must track the constraint because that is the fact the clause was written to fix.

The four-week drill

Week one isolates extraction. The candidate reads discretion clauses and marks only whether a reasonableness qualifier or good-faith duty bounds the power, without yet judging any decision, training the eye to catch the standard before the word "discretion" captures it. Week two adds the validity test: for each clause the candidate states whether a given decision meets the standard, forcing the separation of grant from limit. Week three runs full items under time, selecting answers for the constraint rather than the bare power. Week four mixes bounded discretions with genuinely absolute ones so the candidate cannot assume every grant is qualified, restoring the discrimination that whether a standard bounds it, not whether a discretion is granted, is the fact the question turns on. For the neighboring discipline of reading a classification of one duty as a promise rather than a fact, see the reading representations and warranties versus covenants decoding under the assurance-classification notice guide.

The reader who finishes the drill stops treating every grant of discretion as a free hand and starts reading the discretion notice for the limit it installs. A power to decide can be real and still be bounded, and the candidate who tests the decision against the standard before answering the validity question reads the clause the way it was written — as a specification of a power subject to a constraint, not as a grant of unlimited authority — and stops rewarding a broad word with a freedom it was never meant to confer.