The Crystal and the Cutter: Rough, Cleavage, and Six Centuries of Cut Evolution

Change log: Wave 3 first pass. The A7/A9 readings are among the hardest to fetch in full from the legacy archive; this module teaches the framework and flags title-level sources precisely so nothing pretends to be digested that is not.

Everything about cut starts with a constraint

Diamond has one property that dominates the cutting business and almost no other gemstone shares: perfect octahedral cleavage — it splits along four directions parallel to octahedron faces. Every stage of fashioning, for six hundred years, has been a negotiation with that fact:

The crystal tells you its own shape. Natural diamonds arrive mostly as octahedra, as flattened twins (macles — the “fingerprint” evidence later proves why: growth sectors continue through a twin plane), and as distorted/resorbed pieces. Curved or frosted faces (trigons on the lusterless variety) are growth/resorption signatures, not damage. (The A7 sources: Holmes 1947 documented form/color variation; Seal 1962 read surface structures; Lonsdale’s 1947 single-crystal study is the crystallographic foundation behind it all.)
The cleavage plane tells you how to open it. Historically: sawing against the grain was the expensive, risky operation; cleaving along the plane was the art of the master. The A9 pair Baker 1981 (“Revolution in Diamond Cutting: Laser Sawing”) and the laser-remnant Lab Notes (Allen 2013) bookend the technology: since the 1980s laser drilling/sawing removed the cleavage constraint — leaving, as Allen’s note shows, manufacturing remnants that graders can read — a manufacturing history literally written into finished stones.
Yield beats symmetry. The cutter’s core economics (Caspi 1997’s “Modern Diamond Cutting and Polishing” is the handout’s technical walkthrough): choose the orientation that (a) keeps weight, (b) places inclusions off the table, (c) orients graining for polishing, (d) respects cleavage. When the 1905 Cullinan rough was studied — Scarratt & Shor record the modern examination confirming D color, potentially flawless results — the famous decision tree at Asscher’s was the same algebra: from a 3,106 ct crystal, Joseph Asscher (with brother Abraham Asscher at the Asscher Diamond Co. in Amsterdam) cleaved the stone in February 1908; “could” was never the question; “yield” was.
Rough grading is “begin with the end in mind.” Kautsky’s 2016 GIA research note (link in handout now 404 — see bibliography audit; findable via gia.edu/library) frames the professional discipline: evaluate form, cleavage, inclusions, and then the possible final gems. Any cutter’s mental move — see the finished stone inside the stone — is exactly what a grader reverses when they see a finished stone and deduce its rough (weight retention, symmetry of the plot, graining planes).

Counter script (this is the practical payload): when a client asks why a 1.4 ct stone exists instead of a 1.5, the answer is now professional-grade: “Cutters negotiate four things — weight, clarity placement, polish direction against graining, and cleavage. Sometimes a round 1.40 that dances beats a 1.50 with the table over an inclusion. Rough decides what’s possible; the cutter decides what matters.”

Six centuries of design: the ladder you can walk a client up

Tillander’s 1966 “Six Centuries of Diamond Design” (G&G legacy; link dead — library route) is the citation for the classic sequence every gemology course teaches; learn the logic per step, not just names:

Design	Era (approx.)	Mechanism it solved	What it traded away
Point cut (natural octahedron polished as-is)	1300s–1400s	zero waste; preserves “the most precious of stones” idea (see Vallerano, M01)	almost no brilliance; a shiny crystal
Table / hog’s back	1400s–1500s	flatten the top, show window into interior, kill extinction	more rough lost, small table flash only
Double & single crown	1600s	facets on crown → sparkles (scintillation) and light return begin	yield
Mazarin / Peruzzi (12→16→ crowns+more pavilion)	mid-1600s–1670s	pavilion depth added → real return of light	fire still accidental
Old mine (candlelight)	1700s	high crown, small table, open culet — maximum flash under point-source light	face-up “hole” look under modern lighting
Old European	1800s–1920s	round girdle, more even facets; the first “ideal” debates	still loose tolerances
Modern American round brilliant	1919→	Tolkowsky’s proportional math (58 facets with the culet as #58; GIA’s own 2013 note describes 58-facet brilliants — Pay, in handout A9) → the trade standardizes; GIA’s 2005 cut grade (Module 6) makes “good cut” a measured prediction	nothing — it is what the market now is
Proprietary & native cuts	concurrent	polki and khyzor traditions keep the native form alive (Shor 2016 Polki note — Mogul-Indian table cuts now ‘new fashion’); Watermeyer split-facets (Kerr 1982) and Elara copyright (M13) show the branded-cut game is at least two centuries old	consistency, comparability

That table is the spine of this module. Its punchline for any professional conversation: every “cut” argument in history has been about the same trade — weight vs. light — under whatever light source dominates the era. Candlelight old mines were the “ideal” cut of 1780; daylight brilliants are the ideal of now. A client saying “I love the look of old mine cuts” has said something precise, not nostalgic.

The famous-stones forensics thread (why this history is a science)

Three A9 articles demonstrate that cut-history research has teeth, all in GIA’s journal:

Sunagawa et al. 1998: two polished diamonds from the same rough were matched by growth fingerprinting — internal sector structures are continuous across stones from one crystal. Identity science: a historical pair can be proven. [Source-level annotation; details not yet fetched in full.]
Sucher 2009 (Tavernier Blue) and 2011 (Koh-I-Noor model): crystallographic analysis and laser/X-ray scanning reconstruct how historic stones were cut from what shapes of rough — the Tavernier study even informs cutters’ lost-art conversations. [Title-level this pass.]
Gaillou et al. 2010 (Wittelsbach-Graff & Hope): the persistent legend that both blues came from one rough was tested and rejected with spectroscopic and DiamondView luminescence evidence: distinct dislocation patterns, differing internal strain under crossed polarizers, and distinct phosphorescence decay spectra ruled out a common crystal origin. A professional habit in one citation: when the story is about stones, ask for the stones’ data. (Module 13 completes this thread with the lead cast that DID prove the Hope was recut from the French Blue.)

These are the bibliography’s best argument that “diamond lore” and “diamond science” are not opposites: GIA’s own journal has spent 40 years checking the lore.

What “make” still means on a certificate

Finish the loop to grading vocabulary (full mechanism in M06): proportions (design) → brightness/fire/scintillation (face-up appearance) → polish and symmetry (craftsmanship) — the 2011 G&G symmetry-boundary study (Geurts et al.; abstract read for this course) marks the moment GIA committed to constraining visual judgment with measured boundaries for 10 symmetry parameters, and the diamond-polish/symmetry terminology guide completes the vocabulary. “Made poorly” (1500s) → “cut grade: Excellent” (2005): same human question, four hundred years of instruments between them — the narrative arc from M01, this time in one property.

Self-check

Name the property that dictates sawing/cleaving and its direction.
List the cutter’s four-algebra; give one script that uses it at a counter.
What did the laser (1981 note, 2013 remnant) remove and what did it add to identification?
Why is “old mine = candlelight optimization” a more accurate sentence than “old mine = romantic”?
What can Sunagawa-style fingerprinting prove that appraisal folklore cannot?
Which GIA study rejected a famous “same rough” legend, and with what class of evidence?
Map “make well” → today’s certificate line (three hops).