The standard advice from craft books and graduate seminars runs like this: go to the primary sources. Skip the secondary stuff. Don't let other people's interpretations shape what you'll see. Bring fresh eyes to the original material. The advice has been canonical for so long it feels self-evidently right.
MacCulloch does the opposite. When he wrote his major biographies of Thomas Cranmer and Thomas Cromwell, he went first not to the 16th-century primary sources but to the 17th-century writers who had synthesized those sources before him.1 He read Bishop Gilbert Burnet's eight-volume History of the Reformation and John Strype's many books on the same period — works that "don't mean much to people now, but they were the first people to synthesize all this. They looked at the original sources available to them."1 Only after absorbing Burnet and Strype did MacCulloch go back to the originals.
The reasoning he gives is short and load-bearing: this stops you reinventing the wheel. You see how people in the 17th century thought about the 16th. You get the readers of Burnet, the readers of Strype. You absorb what the field already knows, what it has settled on, what it has framed wrongly — before you go to the originals and try to read them clean.1
The crucial sentence that follows: "Get the cliches in your mind. Do not be dominated by the cliches, but know what they are."1
"Cliches" sounds dismissive, but MacCulloch is using the word in a precise sense. The cliches of a field are its received interpretations — the consensus readings that have been built up by generations of scholars. They are not necessarily wrong. Often they encode genuine pattern recognition by sharp readers working on the material long before you arrived. But they are also the lens through which the field currently sees the primary sources, and that lens shapes what gets seen.
If you go to the primary sources without knowing the cliches, three things happen:
You unwittingly reproduce them. The "fresh reading" you produce is, more often than not, the cliche again, because the cliche became the cliche by being the most obvious reading of the surface material. Without knowing what others have already said, you can't tell whether your reading is original or conventional.
You miss what the cliches encode. Decades of careful scholarship has surfaced features of the material that aren't obvious on first read. A historian going to the Cromwell letters without having read Strype will miss interpretive moves that Strype made in the 17th century and that subsequent scholars built on. The cliche carries information.
You can't see what the cliches obscure. The most important work a historian does is often at the edges of the consensus reading — the places where the cliche is almost right but tilts in a specific direction. You can only see that tilt by knowing what the cliche actually says. Going to the primaries cold means you can't recognize what the field has been getting subtly wrong, because you haven't held the wrong version clearly enough in your mind to identify the tilt.
The cliche-mapping move is, in effect, a calibration step. You absorb the existing interpretive frame so completely that when you go to the primaries, you can recognize where the cliche fits the material and where it stretches it. That recognition is the work the historian is being paid to do.
The second half of MacCulloch's sentence carries the discipline that prevents the move from collapsing into orthodoxy. You get the cliches in your mind. You do not let them dominate. The position you end up in is something like "informed but not captured" — you carry the field's interpretive apparatus with you to the primary sources, but you remain ready to overturn it where the originals demand.
The danger of the move, run badly, is that you absorb Burnet and Strype so thoroughly that you only see the originals through their eyes. The historians MacCulloch is implicitly criticizing — the ones who do reinvent the wheel — are matched by historians who never escape the wheel they were given. Both failure modes are real. The discipline is to be informed enough to know what's been said and free enough to disagree with it when the evidence requires.
How does the historian know when to disagree? MacCulloch doesn't fully spell it out in the interview, but the implied protocol runs through cross-referencing primary against secondary, attending to places where the primaries don't quite fit the synthesized version, and being willing to follow the discrepancy. The cliche tells you what the field expects to see. The primary sources tell you what was actually there. Discrepancies are the historian's leverage — and you can only recognize them if you've absorbed the cliche well enough to spot the misalignment.
There's a side observation in MacCulloch's account that's worth flagging. "It's quite a comforting task because these are very monumental works. Burnet's work runs into eight volumes in its early Victorian edition. Strype similarly wrote endless books. So, read them."1
The throwaway "comforting" is doing real work. The methodology MacCulloch is recommending is not a quick check; it's a substantial commitment of reading time. You're absorbing eight volumes of Burnet before you go to the 16th-century material. That's months of work. But MacCulloch frames the commitment as comforting rather than burdensome — because once you've done it, you know the territory in a way that lets you move confidently through the primaries.
What this implies: the cliche-mapping move trades short-term effort (months of reading what others have said) for long-term confidence (years of clear-eyed work on the primaries). The historian who skips the trade and goes straight to the originals saves time at the front and pays it back over the rest of the project, in the form of slow recognition that "wait, this was already said in 1855."
The protocol is straightforward but disciplined:
Identify the field's canonical secondary works. The synthesizers — the people who took the primary record and produced the first systematic interpretations. For the English Reformation it's Burnet and Strype. For the American Revolution it's Bancroft and the 19th-century constitutional commentators. For Roman history it's Mommsen. Every field has them. Most early-career scholars know they should "be aware of" these works but treat them as historical curiosities. MacCulloch treats them as load-bearing.
Read them properly. Not skim. Not summary-via-Wikipedia. Read the actual monumental work. The interpretations these synthesizers built were built slowly, and the slowness is part of what makes the readings rich. Speed-reading them produces only the surface cliche, not the underlying observations the cliche was built from.
Note the interpretive moves. As you read, mark what the synthesizer is doing with the material. What is he foregrounding? What is he leaving out? What evidence is he leaning on? What is he assuming? These are the cliches, in the precise sense MacCulloch means — the field's settled interpretations and the moves that produced them.
Then go to the primaries. With the cliches in mind, you can read the primary material against them. The places where the primaries fit the synthesizer's frame, you confirm the consensus. The places where the primaries stretch, contradict, or exceed the frame, you have the historian's actual project. The interesting work lives at the discrepancies.
Maintain the "do not be dominated" stance. As you encounter discrepancies, follow them — even when they cut against the synthesized version that has, by now, become familiar. The cliche's job is to give you a baseline against which to detect the new; it is not the answer to the question you're researching.
The strongest case for the methodology is structural: it's how MacCulloch produced biographies of Cranmer and Cromwell that are now standard works in the field. The biographies engage productively with Burnet and Strype throughout, agreeing where the evidence supports them and departing where it doesn't. The departures are visible because MacCulloch had absorbed the older readings. The method's output is itself the evidence.
Tension: the methodology assumes the field has serious secondary literature worth absorbing. For a field with thin or compromised secondary work — recent topics, suppressed histories, areas where the existing scholarship is itself propaganda — the move doesn't apply. The historian has to go to the primaries because the cliches are themselves untrustworthy. MacCulloch's English Reformation field has 400 years of disciplined scholarship to draw on. A historian of recent activist movements may have nothing of the same kind to read first.
A second tension: the methodology privileges which synthesizers count. Burnet and Strype are MacCulloch's choices for the Reformation. Other historians might have started with different synthesizers and produced differently-shaped work. The choice of which cliche to map shapes what discrepancies become visible later. The move is not bias-neutral; the historian inherits the slant of the synthesizer she starts with.
A third tension lives in Karlsson's competing claim ("don't think, look — case studies as still life"). Karlsson's principle is that immersion in the primary material before theorizing is what produces the freshest insight. MacCulloch inverts that order. Both can't be right in their strong forms; the question is when each applies. The implied resolution is that field-thickness matters. Thick fields (extensive secondary literature) reward MacCulloch's move; thin fields (sparse secondary literature) reward Karlsson's. The principle is not "always read secondaries first" or "always read primaries first" but "calibrate to field density."
Open questions:
Run Karlsson's "don't think, look — case studies as still life" through a real project and you can feel the order he's insisting on. You take the specific cases — your friend's failed marriage, the chess game you watched, the awkward dinner — and you stare at them. You don't bring theory to the looking. You let the looking do its work, then the prose forms around what you've actually seen. Trying to do this with Karlsson's order while sitting in MacCulloch's Reformation library would be a disaster. You'd spend years inventing categories Burnet already produced in 1690. Try MacCulloch's order with Karlsson's material and you have a different disaster — you'd read the existing theories about your friend's marriage before paying attention to the marriage itself, and produce prose calibrated to received cliches that have nothing to do with what you actually saw. Both writers reject the casual "do it however" default. They disagree on what comes first because they're working in fields with different amounts of accumulated interpretation behind them. Thick fields reward absorb-then-look. Thin fields reward look-then-think. Diagnose the field's density first; the order follows.
Then bring Tetlock into the room. His superforecasting work landed on something structurally similar from a quantitative angle: base-rate reasoning beats case-specific reasoning. Start with the average outcome for similar situations, then adjust for what's specific to this case. MacCulloch's "read the cliches first" is the qualitative version of the same move. The synthesized account is the base rate. The primary-source departures are the case-specific adjustments. Forecaster and historian, when each is doing her work well, are running the same anchor-then-adjust discipline on different kinds of material.
Susan Orlean's three-phase research-think-write method shares the staged-research instinct but doesn't separate the layers MacCulloch separates. Orlean researches, then thinks, then writes — three phases. MacCulloch reads the synthesizers, then reads the primaries, then thinks, then writes — four phases, because his sources stack two layers deep where Orlean's mostly stack one. The work isn't more elaborate; the material is. Long-dead sources with two centuries of accumulated interpretation between you and the originals warrant more granular staging than primary-witness journalism, where the people are still alive and the synthesizers haven't fully formed yet.
The methodology connects to the literature on expertise acquisition, on knowledge representation, and on the social structure of fields. The shape of the network is: the cliche-mapping move as a specific case of a more general principle that informed-then-not-captured is the optimal stance toward existing knowledge across many domains.
Behavioral Mechanics: Baseline Establishment Protocol (Hughes, Lieberman) — the deception-detection literature emphasizes that you cannot detect deviations without first establishing the baseline. MacCulloch's cliche-mapping is structurally identical at the field level: you cannot detect interesting historiographic moves without first establishing the field's interpretive baseline. The handshake produces a generalizable principle: deviation-detection in any domain requires baseline-establishment first, whether the baseline is a person's normal behavior, a field's consensus reading, or a market's normal volatility. Going straight to the variation, without the baseline, produces unreliable detection.
Eastern Spirituality: Śāstra Tradition and Pramāṇa (or analogous page on textual transmission) — the Indic traditions formalized the principle that primary scripture must be read through the lens of commentary tradition (śāstra) precisely because the commentary tradition encodes the field's interpretive labor across centuries. MacCulloch's move is structurally the same: read the synthesizers (commentary) before the originals (scripture). The convergence is striking. The split is the source of authority: śāstra traditions ground their authority in lineage and realization; MacCulloch grounds his in scholarly discipline. What this means is that the cliche-mapping move is an instance of a much older and more general principle about how traditions transmit interpretive capacity, and that the modern academy's discomfort with this principle (the romantic "fresh eyes" frame) is a specific cultural moment, not a methodological necessity.
Psychology: Expertise and Pattern Recognition (or analogous page) — the expertise literature emphasizes that expert pattern recognition requires extensive exposure to both canonical and edge cases. The novice who memorizes only the canonical patterns can't recognize edge cases when they appear; the novice who skips the canon can't recognize that an edge case is an edge case at all. MacCulloch's methodology trains the historian to recognize both the canon (cliches) and the edges (primary-source discrepancies) by sequencing the exposure correctly. The handshake produces a developmental claim: expertise in historical interpretation has the same structure as expertise in other pattern-recognition fields, and the training sequence MacCulloch describes is the field-specific version of a more general principle about how experts acquire their pattern libraries.
Behavioral Mechanics: Propaganda & Mass Persuasion Hub — propaganda research benefits from the same move. To detect propaganda effectively, you have to first know what the legitimate version of the message looks like — its standard rhetorical moves, its typical sources, its usual structure. Only then can the propagandist's specific deviations from the legitimate template become visible. The handshake produces a practical implication: a journalist or analyst trying to detect propaganda needs to absorb the legitimate genre as thoroughly as the historian absorbs Burnet, so the propagandistic moves become recognizable as deviations. The discipline scales: from individual claim-checking to institutional pattern-detection, the same baseline-then-deviation architecture applies, and analysts trained in this order produce sharper detection than analysts working without the absorbed baseline.
The Sharpest Implication
The methodology contradicts a piece of received wisdom that runs deep in contemporary intellectual culture: the romantic preference for "fresh eyes," for "approaching the material without preconceptions," for "not letting others tell you what to think." If MacCulloch is right, the fresh-eyes stance is the rookie mistake of every field, and the disciplined practitioner specifically does the opposite — saturating themselves in the received interpretation before going to the originals. That's hard to hear, because the fresh-eyes stance has the moral aura of independent thought. But the fresh-eyes stance, in a thick field, produces either accidental reproduction of the consensus (because the consensus reading is the obvious one) or accidental originality that turns out to have been said before. The disciplined move sacrifices the romance of the unmediated encounter for the operational gain of knowing where the field has already been. The third-wire reading is that originality is downstream of saturation, not upstream of it — and that the apparent radical is often the apparent radical because they don't know they're repeating earlier moves.
Generative Questions