Reading for structure
V4’s practice tests whether you can answer questions. This one tests whether you can build the thing the questions are answered from.
Those are different skills, and the second is the one that transfers. A candidate who can answer four questions about a passage they read three times has not learned to read; they have learned to search. The map is what lets you answer from memory the three types that should never require going back, and it is built in one reading or not at all.
So: no options, no keys, no scoring. Model answers to compare against, and a standard to hold yourself to — four phrases, five words each, from one read.
Shared by all four question types below — the anatomy work that every one of them depends on. No options: write your answer, then open the model answers for the set.
| What it does | |
|---|---|
| A1 | States a limit on the evidence — what survives and what it cannot tell you |
| A2 | Qualifies a claim just made — narrows possible to actual |
| A3 | Offers a test case that the preceding argument fails, or an illustration |
| A4 | Decomposes an assumption and reports how each part fared |
| A5 | Explains why the question is hard — and specifies the kind of difficulty |
| A6 | Reports corroboration from independent sources |
| A7 | Restates an objection more precisely — from accuracy to assessability |
| A8 | Bounds the scope of what follows and defers the rest |
What to notice. Not one of the eight required you to know the subject. A1 is about a manuscript, A4 about supply and price, A6 about archaeological deposits — and the function is legible in each without any of that. That is the whole point of reading for structure: the work you do on a first read is subject-independent.
If your answers named the topic — "explains how the technique worked" — you wrote a summary. A summary of A1 would say something about a manuscript. A map says states a limit on the evidence.
| The pivot | What marks it | |
|---|---|---|
| B1 | The correspondence of the principal companies suggests otherwise. | No contrast word. Otherwise carries it, and the attribution shifts: is usually treated gives way to direct assertion |
| B2 | Production figures do not support so simple an account. | No contrast word. A negating verb — do not support — does the work |
| B3 | But preambles are not evidence of purpose… | But, and it is the second half of the sentence that turns, not the paragraph |
| B4 | However, the registers record baptisms and burials, not residence… | However, the most explicit marker in the set |
| B5 | Reefs at similar temperatures nevertheless bleach at very different rates. | Nevertheless, buried mid-sentence rather than opening it |
| B6 | The explanation, however, lies in a reaction… | However, but note the concession before it: is genuine and remarkable |
Two things worth extracting.
First, in B1 and B2 the turn is carried by a verb rather than a conjunction — suggests otherwise, do not support. That is the register V3.3 warns about, and it is where a skimming reader takes the opening position for the author’s own.
Second, in B6 the pivot is preceded by a concession. Its durability is genuine and remarkable grants the strongest version of the claim before narrowing it. When you see a concession, the real point is next.
Roman marine concrete has been described for two centuries as a lost technology, and the description has survived because the structures survive. Harbour works at Baiae and Caesarea stand in seawater after two thousand years, while modern reinforced concrete in comparable settings requires major intervention within fifty.
The comparison is not quite fair, and the reasons are worth setting out. Modern marine concrete contains steel, which corrodes, expands and cracks the matrix from within; the Roman material contains none, and cannot fail in that way. It also cannot span, carry tension, or support the structural forms that reinforcement makes possible. The two materials are asked to do different things.
The durability is nonetheless real and has a mechanism. Seawater percolating through the volcanic ash in the Roman mix drives a slow reaction that precipitates new mineral phases into the pore spaces, so that the material densifies over centuries rather than degrading. Samples drilled from ancient breakwaters contain crystals that formed long after the concrete was placed.
Whether this is usable is a further question. The reaction requires seawater and a particular volcanic ash, and it proceeds over centuries. A material that strengthens over four hundred years and cannot carry tension has a narrow set of applications, and none of them is the set for which modern concrete is used. What the Roman formulation offers is not a replacement but a demonstration that durability and strength are separable properties, which the modern industry has not much treated them as.
Reports that certain corvids manufacture tools have been taken to show a cognitive capacity previously attributed only to primates. The birds strip and shape twigs to extract larvae from crevices, select material of appropriate stiffness, and carry a favoured implement between sites rather than making a new one.
The inference from behaviour to capacity is where the disagreement lies. One position holds that the behaviour is what it appears to be: the bird represents the problem, considers the properties a solution must have, and modifies material to meet them. On this reading the cognitive machinery resembles the primate case and the resemblance is not superficial.
The other position observes that the behaviour is species-typical rather than individual. It appears in birds raised without opportunity to observe it, in nearly the same form, and the material selected is the material the species selects. What looks like problem-solving may be the execution of a highly specific inherited program, no more indicative of general capacity than a spider’s web.
The dispute has been productive in one respect and stuck in another. Experiments designed to separate the accounts — presenting novel problems requiring a modification the species does not naturally make — have produced mixed results that both sides claim. What has not been settled, and may not be settleable by this route, is what would count as a decisive outcome. A bird that solves a novel problem may have general capacity or a broader program; a bird that fails may lack capacity or lack motivation.
Early modern maps are conventionally read as attempts at accurate representation, with their errors treated as failures of survey or information. This reading is anachronistic in a way that obscures what the documents were made to do.
A substantial proportion of surviving maps were produced for litigation, and they were made by and for parties to a dispute. A boundary map commissioned by a manor in a suit over common rights shows the disputed ground at a scale several times that of the surrounding land, marks features that support the commissioner’s case, and omits features that do not. The distortion is not error. It is argument, and contemporaries read it as such.
The consequence for the historian is not that such maps are useless but that they are evidence of a different kind. A map that exaggerates the extent of a waste is poor evidence of the waste’s extent and excellent evidence of what the commissioner wished the court to believe, which is often the more interesting question and is not otherwise recoverable.
None of this applies uniformly. Maps made for navigation were checked against experience and by users with an interest in accuracy, and their errors are generally what they appear to be. The distinction that matters is not between early and modern but between documents whose makers were accountable for accuracy and documents whose makers were not.
It is widely held that the suppression of wildfire over the twentieth century made forests more flammable by allowing fuel to accumulate, and that the severe fires of recent decades are the consequence. The account is broadly correct for the dry pine forests where it was developed, in which frequent low-intensity fire was the historical regime and its removal permitted an unprecedented density of understorey.
It has been extended to forest types where it does not hold. In the wet coastal forests of the Pacific Northwest, the historical fire regime was one of rare, stand-replacing events at intervals of centuries. A century of suppression is a small fraction of that interval and has not appreciably altered fuel loads. Fires in those forests have become more severe, but the mechanism cannot be fuel accumulation, because there has not been time for any.
Where the extension has consequences is in management. Thinning and prescribed burning are appropriate responses to a fuel problem and are being applied in forests that do not have one, at considerable cost and with effects on habitat that are not obviously favourable.
The general lesson is not that the fuel account is wrong. It is that an explanation developed for one system and confirmed within it acquires an authority that does not travel with it, and that the confirmation is often mistaken for generality. The dry-forest evidence is strong. It is evidence about dry forests.
The purple dye of the eastern Mediterranean was produced from the secretions of several marine gastropods, and the industry left middens of crushed shells at sites along the Levantine and Aegean coasts. The scale of production has been estimated from these deposits, and the estimates are very large: tens of thousands of animals for a single garment’s worth of dye.
Those estimates rest on an assumption about yield that has been questioned. They derive from a figure obtained in the nineteenth century by a chemist working with a small sample and a method that has not been replicated. Modern attempts using the same species have obtained yields between four and fifteen times higher, depending on the season of collection, the size of the animal and the handling of the secretion before it is exposed to light.
If the higher yields are representative, the middens imply an industry an order of magnitude smaller than has been supposed, with consequences for what the dye cost and who could obtain it. If the lower figure is right, the received account stands.
The archaeological evidence does not decide between them, and it is worth being clear why. Midden volume measures animals processed, not dye obtained, and the ratio between them is precisely what is in question. Settling it requires replication of the ancient method, which is not recorded, under the conditions in which it was practised, which are not known.
That human sleep was historically biphasic — a first and second sleep separated by an hour or two of waking — is now widely repeated, on the strength of references to "first sleep" in early modern English sources and a laboratory study in which subjects deprived of artificial light settled into a segmented pattern within weeks.
The literary evidence is thinner than its currency suggests. The phrase appears in a small number of texts, mostly in contexts where a character is woken rather than waking naturally, and comparable phrases in other languages have been collected without much attention to whether they describe a norm or an interruption. That a period has a word for waking in the night is consistent with waking in the night being ordinary and also with its being notable.
The laboratory evidence is stronger but narrower. It establishes that under fourteen hours of darkness a segmented pattern emerges, which is a fact about photoperiod rather than about history. Whether pre-industrial populations experienced fourteen hours of darkness depends on latitude, season, and what firelight and rushlight amounted to, none of which the study addresses.
The claim may still be true. What it lacks is evidence of the kind that would establish it, and the confidence with which it now circulates has outrun the support. It is a good example of a finding that is plausible, well publicised, and resting on two sources neither of which was designed to answer the question it is being used to answer.
| Exercises | ||
|---|---|---|
| Layer 1 · paragraph function | 8 | |
| Layer 2 · find the pivot | 6 | |
| Layer 3 · full map, six passages × 4 | 24 | |
| Total | 38 | |
| Drills | Items | |
| --- | ---: | ---: |
| Sentence Correction | 90 | 66 |
| Critical Reasoning | 46 | 66 |
| Reading Comprehension — V3 map work | 38 | — |
| Reading Comprehension — V4 questions | 6 | 62 |
| Total | 180 | 194 |
Map. ¶1 the received description · ¶2 the comparison is unfair · ¶3 the durability is real, with a mechanism · ¶4 whether it is usable, and what it actually shows
Pivot. "The comparison is not quite fair." No contrast word; the negation carries it. Note that the passage pivots twice — again at "The durability is nonetheless real" — which is common in passages that both qualify and then partially restore a claim.
Stance. The author accepts that Roman marine concrete is genuinely durable and that the mechanism is understood, but rejects the framing of it as a lost superior technology, and does not claim it has practical application today — while allowing that it demonstrates something the modern industry has neglected.
Shape. Claim qualified. The identifying structure is that the opening claim is neither accepted nor rejected: it is narrowed twice and then given a different significance.
Map. ¶1 the behaviour and what it has been taken to show · ¶2 first reading: genuine capacity · ¶3 second reading: inherited program · ¶4 why the dispute is stuck
Pivot. There is no single pivot, and noticing that is the exercise. The passage does not challenge a position; it lays two out. The sentence that organises it is "The inference from behaviour to capacity is where the disagreement lies" — which is a signpost rather than a turn.
Stance. The author takes no position on which reading is correct and holds that the experimental route so far pursued may be incapable of settling it, because no outcome has been specified that would count as decisive.
Shape. Two positions compared. Identified by the symmetry: one paragraph each, neither given the last word, and a fourth paragraph about the dispute rather than about the birds.
Map. ¶1 the conventional reading, rejected as anachronistic · ¶2 what litigation maps were for · ¶3 what follows for the historian · ¶4 the limit of the argument
Pivot. "This reading is anachronistic in a way that obscures what the documents were made to do." Second sentence of the passage — early, direct, and with no contrast word.
Stance. The author holds that many early modern maps should be read as argument rather than as failed representation, and is explicit that this does not apply to all of them: the operative distinction is accountability for accuracy, not date.
Shape. Old view challenged, with an unusual feature — ¶4 limits the author’s own claim rather than the one being challenged. A passage that qualifies itself in the last paragraph is stating a narrower position than it appeared to be stating, and the stance question will turn on that.
Map. ¶1 the fuel account, correct where developed · ¶2 where it does not hold · ¶3 the management consequence · ¶4 the general lesson
Pivot. "It has been extended to forest types where it does not hold." Paragraph-initial, no contrast word, and the whole turn is in the phrase where it does not hold.
Stance. The author accepts the fuel account for dry forests, rejects its extension to wet forests, and — the part that is easy to miss — is making a general point about how confirmation within one system gets mistaken for generality.
Shape. Claim qualified, though it reads at first like old view challenged. The difference matters: ¶1 endorses the account explicitly, and ¶4 says the fuel account is not wrong. A reader who took ¶1 as a view about to be demolished will answer a stance question incorrectly.
Map. ¶1 the received estimates of scale · ¶2 the assumption they rest on, questioned · ¶3 what turns on it · ¶4 why the evidence cannot settle it
Pivot. "Those estimates rest on an assumption about yield that has been questioned." Paragraph-initial, no contrast word, and it is the phrase rest on an assumption that marks the turn — a claim described as resting on something is about to have that something examined.
Stance. The author holds that the received estimate of the industry’s scale depends on an unreplicated figure, and does not claim the higher modern yields are correct — the position is that the question is open and that the archaeological evidence is structurally unable to close it.
Shape. Claim qualified, ending in an unresolved state. Note ¶3’s symmetry — if the higher yields … if the lower figure — which is a reliable sign that the author is not going to choose.
Map. ¶1 the claim and its two supports · ¶2 the literary evidence is thin · ¶3 the laboratory evidence is narrow · ¶4 the claim may be true but is undersupported
Pivot. "The literary evidence is thinner than its currency suggests." Paragraph-initial, and the comparison thinner than its currency does the work of a contrast word.
Stance. The author does not deny that sleep may historically have been biphasic, and holds that neither line of evidence supports the claim as confidently as it is now made — a position about the quality of the support, not about the truth of the claim.
Shape. Old view challenged, with the same self-limiting move as Passage 3: "The claim may still be true." Options on a stance question that read the author as rejecting biphasic sleep will be wrong, and they will be tempting.
Three of the six passages qualify their own argument in the last paragraph. Passages 3, 4 and 6 all end by narrowing what the author is claiming, and in each the stance is more modest than the first three paragraphs suggest. That is the error V3.4 names — reading the author as more committed than the text supports — and this block is weighted toward it deliberately.
Four of the six pivots have no contrast word. Is not quite fair, has been extended, rest on an assumption, thinner than its currency. If you found fewer than four, the habit to build is the one from V3.3: watch for a verb that negates, and watch for the attribution to stop.
Passage 2 has no pivot at all, and if you looked for one you learned something useful. Not every passage challenges a position. Some lay two out and decline to choose, and forcing a pivot onto one of them produces a map that misdescribes the whole thing.