Level 5 — Advanced · Technical & Conceptual Reading
Lesson 72: Advanced Reading of Argument Structure: Premises, Evidence, Inference & Conclusion
How to take a single argument apart into its working parts — premise, evidence, inference, hidden assumption, and conclusion — and catch the gap where an author's conclusion quietly claims more than the evidence actually supports.
Learning Objectives
By the end of this lesson, you should be able to:
- Separate a premise from its supporting evidence and from the argument's final conclusion, reading 'this suggests' and 'therefore' as the two structural bridges connecting them
- Distinguish genuine evidence from an example, a case study, an analogy, or an appeal to authority — none of which, by itself, proves a general claim
- Identify an argument's gap — the missing bridge between premise and conclusion — and its hidden assumption, distinguishing a necessary assumption (without which the argument collapses) from a merely helpful one
- Catch scope shift and time shift as common argument weaknesses, and read 'because' and 'therefore' as signals of a causal or logical claim rather than proof that the claim is valid
- Rate an argument's overall strength — strong, moderate, weak, or unsupported — by checking whether its evidence, assumptions, and scope actually match the weight of its conclusion
Introduction
Lesson 71 tracked how an author's claim develops across a whole passage — broad opening, evidence, complication, refined final position. Today's lesson zooms into a single argument and takes it apart at the joints: which statement is the premise, which is evidence, which is an inference drawn from that evidence, and which is the conclusion the whole thing is actually building toward. This is the core skill of critical reading — not just following an argument, but being able to point to exactly which piece is doing which job.
The Basic Structure of an Argument
Premise → evidence/support → inference → conclusion
Employees who receive structured training make fewer billing errors. The new system requires employees to interpret more detailed billing data. Therefore, organizations implementing the system should provide additional training. Premise: employees who receive structured training make fewer billing errors. Supporting fact: the new system requires employees to interpret more detailed billing data. Inference: more interpretation demand → training becomes more important. Conclusion: organizations should provide additional training. Four distinct jobs, four distinct sentences.
What Is a Premise?
The statement a conclusion is actually built on
A premise is whatever answers the question "why should I believe the conclusion?" The port should upgrade its network infrastructure because current equipment frequently causes communication failures. Conclusion: the port should upgrade its network infrastructure. Premise: current equipment frequently causes communication failures.
Not Every Reason Is an Equally Good Premise
Count the reasons, but also weigh their quality
The system is expensive, difficult to maintain, and unpopular among some employees. Therefore, it should not be adopted. Three reasons are offered — cost, maintenance, popularity — but they're not equally strong. Cost can be an objective, measurable factor. Maintenance difficulty can be an operational criterion. Unpopularity is preference-based, and preferences alone rarely settle a question this consequential. Reading an argument well means evaluating each reason's evidential quality, not just tallying how many reasons were listed.
Evidence Versus Premise: A Layered Structure
A premise can itself rest on evidence
Employee training improves accuracy functions as a premise. A three-year study of 15,000 employees found that structured training reduced processing errors by 23% is evidence supporting that premise.
EVIDENCE
↓
SUPPORTS
↓
PREMISE
↓
SUPPORTS
↓
CONCLUSION
Read this as a layered structure — evidence holding up a premise, which in turn holds up a conclusion — not a flat, single-level chain.
Evidence → Claim → Conclusion: A More Realistic Chain
Evidence rarely jumps straight to a recommendation
A study found that automated inventory systems reduced stock discrepancies by 18%. This suggests that automation can improve inventory accuracy. Therefore, organizations with high levels of inventory error may benefit from automation. Evidence: 18% reduction in stock discrepancies. Interpretation: automation can improve inventory accuracy. Recommendation: organizations may benefit from automation. Evidence doesn't lead directly to a recommendation — an interpretation step sits in between, and skipping past it is one of the most common ways an argument gets misread as stronger than it is.
"This Suggests": The Inference Bridge
The phrase that carries you from observed fact to interpretation
Processing errors declined after employees received additional training. This suggests that employee capability was an important factor in the improvement. Observed fact: errors declined. Inference: employee capability was important. The inference is not itself an observation — it's a step the author is asking you to take, one that should be checked rather than absorbed automatically.
"Therefore": The Conclusion Bridge
The phrase that carries you from premise to conclusion
Employee training improved data interpretation. Therefore, training should be included in the implementation plan. First statement: premise or reason. "Therefore": inference marker. Second statement: recommendation or conclusion. Two bridges, two different phrases — "this suggests" typically carries evidence to interpretation; "therefore" typically carries premise to conclusion.
Reason Markers: Because, Since, Given That, Considering That
Four ways of introducing the 'why' behind a claim
The company delayed deployment because the security review was incomplete.
WHY?
↓
Security review incomplete
↓
THEREFORE
↓
Deployment delayed
Since the available data are incomplete, the results should be interpreted cautiously — "since" here means "because"; watch context carefully, since "since" can also mark time (the system has improved since the update was deployed). Given that the sample was relatively small, the findings should be interpreted cautiously foregrounds a premise formally. Considering that maintenance costs have increased, the company should review the system's long-term value works the same way — a formal way of flagging the consideration driving a recommendation.
Basis Markers: Based On, On the Basis Of, In Light Of
Naming what a conclusion rests on
Based on the available evidence, the researchers concluded that the intervention had limited effects.
AVAILABLE EVIDENCE
↓
BASED ON
↓
RESEARCHER CONCLUSION
On the basis of these findings, the committee recommended further testing is a more formal version of the same move, common in academic and policy documents. In light of the recent failures, the organization decided to revise the procedure signals that new, relevant information influenced a decision.
"Accordingly" and "Consequently"
Signalling action or result flowing from prior reasoning
The existing process was found to be unreliable. Accordingly, a new verification procedure was introduced — unreliable → accordingly → new procedure: action following from a stated finding. The input data were incomplete. Consequently, several calculations had to be repeated — cause → consequence. Both can signal a causal relationship, but as with every connector in this lesson, verify the author's actual reasoning rather than trusting the connector word alone.
"This Means That…"
Drawing out a consequence or interpretation
The system requires manual verification for exceptional cases. This means that full automation is not currently possible. The inference — full automation isn't currently possible — is drawn directly from the stated requirement (manual verification for exceptions), not introduced as a separate, independently supported fact.
"In Other Words" Is Not New Evidence
Connecting back to Lesson 71's restatement markers
The process is highly dependent on data quality. In other words, poor input data can produce unreliable results. The second sentence restates the first — it is not new evidence, and shouldn't be counted as independent support when you're weighing how much backing an argument actually has. Restatement ≠ independent support.
Example Is Not Evidence
An illustration is not proof of a general claim
Many successful companies use automation. For example, Company X automated its billing process. Company X's example is offered — but "one company succeeded, and it used automation" does not establish that automation caused that success. An example illustrates a claim; it doesn't, by itself, prove one.
Case Study Is Not General Proof
What worked in one place doesn't automatically generalise
A port implemented a digital tracking system and reduced truck waiting times by 30%. Fair conclusion: the system worked in that port, under those conditions. Unsupported leap: the system will reduce waiting times by 30% in every port. This is a scope mismatch — the case study's actual scope (one port, one set of conditions) is narrower than the generalised claim being built on top of it.
Analogy Is Not Evidence
Connecting back to Lessons 59 and 60 — an analogy builds intuition, not proof
Managing data is like managing cargo: both require careful tracking. This can help a reader understand a concept — but the analogy itself is not empirical evidence for any claim about data management's actual effectiveness.
Authority Is Not Automatic Proof
An expert's opinion is a claim, not proof of that claim
A leading expert argues that AI will transform logistics. An expert opinion has been offered — but "an expert said it" does not automatically make a claim true. The critical question remains: what evidence does the expert actually provide for this specific claim?
Hidden Inference
A conclusion the author expects you to draw, without stating it outright
The new system requires employees to learn several additional procedures. Training time has already increased substantially during the pilot phase. The author never states outright: implementation will require additional training resources. But the inference is clearly invited: more procedures + increased training time → additional training resources likely needed. Reading well means noticing when a conclusion is implied rather than stated, and being able to state it explicitly yourself.
The Argument Gap
The missing bridge between premise and conclusion
The software is widely used. Therefore, it is the best software for our organization. Gap: "widely used" → "best for our organization" — this relationship is never actually established. The missing premise that would close the gap: the needs of our organization are similar to those of organizations using the software. That missing piece is the argument's hidden assumption, and naming it explicitly is often the single fastest way to evaluate whether an argument actually holds up.
Assumptions, Revisited
Building on this course's earlier work on assumptions
Automation should be adopted because it reduces labor costs. Hidden assumption: reducing labor cost is sufficiently important to justify adoption. Automation should be adopted because competitors are using it. Hidden assumption: competitors' adoption indicates that automation is appropriate for our organization too. Every argument with a gap has an assumption filling that gap, whether the author states it or not.
Necessary Assumption Versus Helpful Assumption
One kind of assumption an argument can survive without; one it can't
A helpful assumption makes an argument stronger, but the argument can survive without it. A necessary assumption is one the argument collapses without — if it's false, the conclusion no longer follows. The company should adopt the software because it reduces processing time. Necessary assumption: processing time is an important criterion for this company's objective. If processing time turns out to be essentially irrelevant to the company's actual goals, the conclusion falls apart — that's what makes this assumption necessary rather than merely helpful.
Testing an Argument: "If This Is True, Then…"
A powerful, simple test for whether a premise actually supports its conclusion
Ask: if the premise is true, does the conclusion automatically follow? The system is faster → the system is better. Better in what respect? If the relevant criterion really is speed, the argument holds up reasonably well. If the relevant criteria also include reliability, cost, security, and maintainability, speed alone isn't enough to support "better" — the argument is weaker than it first appears.
Missing Criteria Weaken an Argument
Connecting back to Lesson 65's evaluation framework
Solution A is better than Solution B. Critical questions: better in terms of what — cost, speed, reliability, security, scalability, usability? An argument built on an unqualified "better" is exactly the incomplete evaluation Lesson 65 trained you to catch — here, applied to judging the argument's own soundness.
Scope Shift
A conclusion quietly claims a broader scope than its premise actually covers
Most employees prefer the new system. Then: therefore, the new system is better. Scope shift: preference → overall quality. What was the preference actually based on? Maybe ease of use — in which case, usability is supported, but reliability, security, and cost remain completely unaddressed. The conclusion has quietly expanded past what the premise actually established.
Time Shift
Connecting back to Lesson 62 — sequence in time is not causation
Performance improved after the new system was introduced. Conclusion: the new system caused the improvement. Problem: after ≠ because of. Training, staffing changes, process redesign, or seasonal variation could plausibly have occurred around the same time — the time shift from "improved after X" to "caused by X" is exactly the sequence-versus- causation gap Lesson 62 first built.
The Correlation-to-Causation Gap
Connecting back to Lesson 56 — several explanations can fit the same association
Employees who use the new system perform better. Possible explanations: the system genuinely improves performance; better employees are simply more likely to use the system; trained employees use the system more; management selectively assigns the system to already high-performing teams. Association ≠ causation — one of the most frequently violated principles in argument writing, precisely because several very different explanations can produce the identical observed pattern.
"Therefore" Doesn't Certify Validity
Connecting back to Lessons 64 and 71
The system is expensive. Therefore, it is ineffective. "Therefore" is present — but the reasoning gap (expensive → ineffective) is never actually closed. A system can easily be expensive and highly effective, or expensive and cost-effective. "Therefore" signals the author believes a conclusion follows; it never certifies that it actually does.
"Because" Doesn't Certify Causal Truth
Connecting back to Lesson 62's causal-strength ladder
The project failed because employees resisted the change. This is the author's causal claim — not a proven fact. A critical reader asks what evidence supports it, and considers alternatives: poor implementation, inadequate training, technical problems, unrealistic deadlines, budget limitations. "Because" signals a causal claim; it does not prove the claim.
Rating Argument Strength
Four levels for judging how well an argument holds together
A Full Worked Example
Evidence, scope, qualification, inference, limitation, and a carefully qualified conclusion
A six-month pilot found that automated billing reduced average processing time by 28%. The reduction was observed across four operational teams, although the teams differed in workload and staffing levels. These findings suggest that automation can improve processing efficiency. However, the pilot did not measure long-term maintenance costs or the effect of employee training. Therefore, while the results provide support for wider implementation, they do not by themselves establish that full-scale deployment would be cost-effective.
Evidence: 28% reduction. Scope: four teams. Qualification: teams differed in workload and staffing. Inference: automation can improve processing efficiency. Limitations: long-term maintenance not measured; training effect not measured. Conclusion: evidence supports wider implementation. Qualification of the conclusion: cost-effectiveness not established. The author never claims "automation is definitely cost-effective" — only that evidence supports wider implementation, while cost-effectiveness specifically remains unestablished. That's a genuinely qualified conclusion, not an overreach.
Argument Compression
Connecting back to Lessons 47, 63, and 71's compression skill
The paragraph above compresses to one sentence without losing its real shape: a multi-team pilot found meaningful processing-time improvements from automation, supporting further implementation, but insufficient evidence exists to establish its long-term cost-effectiveness. Notice what survives: evidence, scope, benefit, recommendation, and limitation — exactly the parts a careless summary drops first.
An Argument Tracking Template
A note-taking structure for any chapter-length argument
An Advanced Reading Algorithm
Eight steps for any argument-heavy paragraph
Master Argument Model
The full architecture behind a fully developed argument
Vocabulary in Context
premisenounformal
a statement offered as the reason a conclusion should be believed (আশ্রয়বাক্য/পূর্বশর্ত)
Often with a false premise, underlying premise, accept the premise
“The argument's premise, that all delays are avoidable, is doubtful.”
argument gapnoun phrasecourse term
the missing logical bridge between a premise and a conclusion that the premise alone does not establish (যুক্তির শূন্যতা)
Often with an argument gap, spot the argument gap, fill the argument gap
“The report has an argument gap: it shows costs rose but never shows the change caused them.”
necessary assumptionnoun phrasetechnical
an unstated premise without which an argument's conclusion no longer follows, as distinct from one that merely strengthens it (আবশ্যিক অনুমান)
Often with identify the necessary assumption, a necessary assumption, hidden necessary assumption
“The plan's necessary assumption is that suppliers will deliver on time; without it, the schedule fails.”
scope shiftnoun phrasecourse term
a conclusion that quietly claims a broader scope than its supporting premise actually covers (পরিধির স্থানান্তর)
Often with commit a scope shift, subtle scope shift, scope shift from
“A scope shift occurs when data on one terminal is used to draw conclusions about every terminal.”
hidden inferencenoun phrasecourse term
a conclusion an author invites the reader to draw without stating it directly (অন্তর্নিহিত অনুসিদ্ধান্ত)
Often with draw a hidden inference, make a hidden inference, hidden inference behind
“The advert's hidden inference is that competing terminals are unsafe, though it never says so.”
argument strengthnoun phraseformal
a rating of how well an argument's evidence, assumptions, and scope actually support its conclusion, from unsupported to strong (যুক্তির শক্তি)
Often with assess argument strength, judge argument strength, argument strength of
“Argument strength depends on how much of the conclusion the evidence actually covers.”
case studynoun phraseneutral
a report of what happened in one specific instance, which supports a claim about that instance but not automatically a general claim (কেস স্টাডি)
Often with a case study, a case study of, use a case study
“The chapter uses a case study of one Asian port to illustrate a wider trend.”
inference bridgenoun phrasecourse term
a phrase such as 'this suggests' or 'therefore' that carries the reader from evidence to interpretation (অনুসিদ্ধান্ত সেতু)
Often with an inference bridge, use an inference bridge, inference bridge such as
“'This suggests that' is an inference bridge from the data to the author's interpretation.”
Guided Reading Practice
Read this passage once, straight through. Then identify its premise, its evidence, its inference, its hidden assumption, and its final qualified conclusion, before checking your reading against the notes that follow.
Warehouses that adopted barcode scanning reported fewer picking errors than those relying on manual entry, based on a survey of 40 mid-sized distribution centers. This suggests that scanning technology can reduce human error in order fulfillment. However, the surveyed warehouses also tended to have more experienced staff and lower turnover than the industry average, so staffing stability may have contributed to the difference as well. Because the survey did not control for staff experience, it would be premature to conclude that barcode scanning alone accounts for the lower error rates. Therefore, while the association supports further investigation, it does not by itself justify treating scanning technology as a guaranteed fix for picking errors.
Premise/evidence: warehouses with barcode scanning reported fewer picking errors, based on a 40-center survey. Inference bridge, "this suggests": scanning technology can reduce human error. Competing explanation surfaced: surveyed warehouses also had more experienced, stable staff — a confound the passage names explicitly. Explicit scope-check: "because the survey did not control for staff experience, it would be premature to conclude" — directly blocking the correlation-to-causation leap this lesson has repeatedly warned about. Final qualified conclusion: the association supports further investigation, but does not justify treating scanning as a guaranteed fix — a considerably narrower claim than the opening sentence might suggest on its own.
Golden Rule
Golden Rule
Never jump directly from evidence to conclusion — always ask: what does the evidence actually show, what is the author inferring from it, what assumption connects that inference to the conclusion, and is the conclusion stronger than the evidence actually earns?
Lesson Summary
This lesson took a single argument apart into its working parts. You separated premise from evidence from inference from conclusion, and practised reading "this suggests" and "therefore" as two distinct structural bridges rather than interchangeable connectors. You learned that an example, a case study, an analogy, and an appeal to authority are none of them, by themselves, genuine evidence for a general claim. You found the gap between a premise and an overreaching conclusion, and named the hidden assumption filling that gap — distinguishing a necessary assumption (without which the argument collapses) from a merely helpful one. You caught scope shift and time shift as two recurring argument weaknesses, and practised treating "because" and "therefore" as signals of a claim, never as proof the claim is sound. You closed by rating a full worked argument's strength and compressing it into one sentence without losing its evidence, scope, or qualification.
The core question this lesson trains you to ask
Every argument you read from now on deserves the same question: what exactly supports the author's conclusion, and what part am I merely being asked to assume? Holding that question in mind, sentence by sentence, is the difference between following an argument and actually evaluating one.
Practice: Test What You've Learned
Work through every question yourself before checking anything.
Before you start
For every argument below, label its premise, its evidence (if any), and its conclusion separately — "this is the argument" is not yet a complete answer.
Part A — Premise, Evidence, and Inference
- The clinic should extend its hours because patient wait times have increased significantly this year. Identify the premise and the conclusion.
Show answer
Premise: patient wait times have increased significantly this year. Conclusion: the clinic should extend its hours. The premise, introduced by "because," is offered as the reason to believe the conclusion. - A survey of 200 patients found that 80% preferred online booking. This suggests that expanding online booking could improve patient satisfaction. Identify the evidence and the inference drawn from it.
Show answer
Evidence: a 200-patient survey found 80% preferred online booking. Inference, carried by "this suggests": expanding online booking could improve patient satisfaction. The inference is a step beyond the raw survey result, not the observed fact itself. - Explain, using this lesson's terms, why evidence rarely leads directly to a recommendation without an interpretation step in between.
Show answer
Raw evidence, such as a measured percentage or reduction, only describes what was observed in a specific study or sample. An interpretation step, marked by phrases like "this suggests," is needed to turn that observation into a general claim, and only then can a recommendation, marked by "therefore," reasonably follow — skipping the interpretation step is what makes an argument look stronger than its evidence actually supports.
Part B — What Is Not Evidence
- Many top-performing schools have adopted project-based learning. For example, School X saw rising test scores after switching. Explain why School X's case is an example rather than proof that project-based learning caused the improvement.
Show answer
School X's case only illustrates that one school switched to project-based learning and later saw rising scores; it does not rule out other explanations, such as changes in staffing, funding, or student population, that might have driven the improvement instead. An example can make a claim vivid, but it cannot, by itself, establish that the method caused the result. - A well-known economist claims that remote work will permanently reshape city planning. Explain why this claim, on its own, is an appeal to authority rather than evidence.
Show answer
The sentence tells us who made the claim, not what supports it. An expert's status makes the claim worth taking seriously, but it does not itself constitute data, a study, or reasoning that could confirm the prediction — the critical question is always what evidence the economist actually offers for this specific claim. - Running a company is like piloting a ship: both require careful navigation through changing conditions. Explain why this analogy cannot serve as evidence for any specific claim about company management.
Show answer
An analogy highlights a resemblance between two different things to help a reader understand a concept intuitively; it does not report any actual data about companies or their outcomes. Because running a company and piloting a ship are genuinely different activities in most respects, the comparison cannot prove any specific, factual claim about how companies should actually be managed.
Part C — Argument Gaps and Assumptions
- The app has the highest number of downloads in its category. Therefore, it is the best app for our team's needs. Identify the argument gap and state the missing premise that would close it.
Show answer
The gap is between "highest number of downloads" and "best for our team's needs" — popularity is never shown to relate to fit for this specific team. The missing premise that would close the gap: the team's needs are similar to those of the general population of users who downloaded the app. - We should switch vendors because our main competitor already has. Identify the hidden assumption this argument depends on.
Show answer
The hidden assumption is that the competitor's decision to switch vendors is a reliable indicator that switching is appropriate for this company too — that what is right for the competitor's situation is also right for this company's situation. - A company adopts a new tool because it reduces meeting time. Identify a necessary assumption this argument depends on, and explain what happens to the conclusion if that assumption turns out to be false.
Show answer
The necessary assumption is that reducing meeting time is an important goal worth pursuing for this company. If that assumption turns out to be false, for example if meeting time was never a significant problem for the company, the conclusion that the tool should be adopted no longer follows, since the argument's entire reason for adopting it collapses.
Part D — Scope Shift, Time Shift, and Causal Claims
- Customer complaints dropped after the new support software was introduced. Therefore, the software caused the drop in complaints. Identify the type of shift this argument commits, and name one alternative explanation.
Show answer
This is a time shift: it treats "the drop happened after the software was introduced" as proof that the software caused it, when sequence in time is not causation. An alternative explanation: staff received additional customer-service training around the same period, and that training, not the software, could have driven the improvement. - Most survey respondents said they enjoyed the new office layout. Therefore, the new layout improves productivity. Identify the scope shift in this argument.
Show answer
The scope shift moves from enjoyment or preference to productivity, two different things entirely. The survey only establishes that employees liked the layout; it says nothing about whether that enjoyment actually translates into measurable output or efficiency gains. - The redesign was expensive. Therefore, it was not worth doing. Explain why "therefore" here does not certify that the reasoning is valid.
Show answer
"Therefore" signals that the author believes a conclusion follows, but the actual reasoning gap between "expensive" and "not worth doing" is never closed. A redesign can easily be expensive and still worth doing, if its benefits, such as increased revenue or reduced long-term costs, outweigh that expense — cost alone does not settle the question of value.
Part E — Full Passage Analysis
Return to this lesson's guided reading passage about barcode scanning in warehouses.
- State the passage's evidence and its inference bridge in your own words.
Show answer
Evidence: a survey of 40 mid-sized distribution centers found that warehouses using barcode scanning reported fewer picking errors than those relying on manual entry. Inference bridge, "this suggests": scanning technology can reduce human error in order fulfillment. - Identify the competing explanation the passage raises, and explain why it matters.
Show answer
The competing explanation is that the surveyed warehouses also had more experienced staff and lower turnover than average, so staffing stability, not the scanning technology itself, may have contributed to the lower error rates. It matters because it identifies a confound the survey did not control for, which means the technology's effect cannot be cleanly separated from the staffing difference. - Identify the sentence that explicitly blocks a correlation-to-causation leap, and explain how it does so.
Show answer
"Because the survey did not control for staff experience, it would be premature to conclude that barcode scanning alone accounts for the lower error rates" blocks the leap. It does so by naming the exact missing control, staff experience, and explicitly refusing to credit the technology alone with the entire observed difference. - Explain why summarising this passage as "barcode scanning reduces picking errors" would overstate what the author actually established.
Show answer
That summary presents scanning as a proven, sole cause of the improvement, but the author only claims the association supports further investigation, given the unaddressed staff-experience confound. It ignores the explicit qualification that the evidence does not by itself justify treating scanning as a guaranteed fix for picking errors, which is the author's actual, narrower conclusion.
Next: Lesson 73 — Advanced Reading: Hidden Assumptions, Missing Links & Argument Gaps: a deeper, more systematic pass on the argument gap — a named taxonomy of gap types and a concrete test for which assumptions actually matter.