Nankin-Kanna

Identity

Name. Nankin-kanna (南京鉋; なんきんかんな), also voiced nankin-ganna by rendaku (the regular sequential-voicing shift kanna → ganna in compounds). The name is carried by mainstream lexicography this time — 精選版 日本国語大辞典 (via Kotobank) gives 南京鉋 / なんきんかんな and defines it: "反り台鉋の一つ。円溝を削るのに用いる小形のもの" — "a type of sori-dai-kanna (curved-sole plane); a small one used for cutting circular grooves" [N1]. A precision note worth keeping: the dictionary's defining job — cutting circular grooves (円溝) — is narrower than the broad convex/concave curve-finishing that living furniture-makers and the toolmakers describe [N6][N10]. Whether the dictionary preserves an older, narrower sense or simply compressed the tool to one job is unresolved — flagged, not collapsed. An alternate name exists and carries a dissent: the expert Kurayuki Abe argues the "meaningless" 南京 should be dropped for the descriptive 棒鉋 (bō-kanna, "rod/stick plane") on morphological grounds [N7] — a named practitioner's objection to the very name, recorded as texture.

Why "Nankin." 南京 is Nanjing, the Chinese city — but as a prefix in Japanese it is a productive label for things that are foreign-arrived, Chinese-associated, or small/novel/exotic. The National Diet Library reference database, citing the Kurashi no Kotoba Shin Gogen Jiten (Kodansha, 2008), records that for things from abroad the old prefix kara (唐) gave way in the early-modern period to nankin — 南京豆 (peanut), 南京虫 (bedbug), 南京更紗 (Nanjing cotton), 南京錠 (padlock, "a lock of foreign origin") [N2]; デジタル大辞泉 gives the prefix's two senses outright: ① Chinese-produced or Chinese-arrived, ② rare, or small-and-cute [N3]. So the name reliably tells us the tool was perceived as foreign/Chinese-style — but not that it physically came from Nanjing. Odate himself only ventures "probably named after a city in China… where they probably originated" [N16]; held as marked-foreign, origin-city UNVERIFIED.

Family / category. Surfacing tools → bodied curved-work planes → the spokeshave branch, in its Japanese form. It is the functional sibling of the Sen (the sole-less draw-scraper) for repeatable curved profiles — a structural counterpart Hewn compares it to, not a documented historical successor — the curved-work counterpart of the flat Kanna (dai-kanna) in the same bodied-plane family, and the cross-tradition twin of the Western Spokeshave — independent convergence on "a small body with two handles and a blade in the middle, for shaping curves." Crucially, and unlike the Sen and Yariganna, it IS a bodied plane: it has a dai, a sole, and a mouth — and the sole is the whole point.

One line. A small two-handled bodied plane — a laminated hagane/jigane blade wedged in a short white-oak body, the sole left slightly curved (and shaped to match the work), a wing handle projecting from each side — drawn or pushed across convex, concave, and round wood to fair and finish a curved surface a flat-soled plane cannot ride; the Japanese spokeshave, whose name marks it as Chinese-associated, and the bodied member that closes the surfacing-tool family [N1][N6][N10].

✦ How NOT to use it

The heaviest section. Note up front: like the Sen, the Yariganna, and the whole saw root, the nankin-kanna's danger profile is thin on DOCUMENTED injury — no first-person nankin-kanna laceration was found this run — but the hazards are structurally clear (an exposed blade on a short sole, worked close to the body) and are documented for its cousins the Spokeshave and Drawknife; that material is used as mechanism-by-analogy and labelled as such. Where this run's outside review found specific hazard claims too speculative or too precise given the lack of nankin-kanna-specific incident evidence, the guidance below has been replaced with conservative, general edged-tool precautions.

not-for (jobs it should never do):

• Flat panels and faces to a calibrated thickness. Its curved, short sole cannot reference a flat surface — it will leave a faceted, rounded result. Truing and thicknessing flats is the flat Kanna's job; the nankin-kanna is a curved-work specialist [N6][N5, mechanism].
• Long straight jointed edges. Too short a sole to register a straight edge over length; it introduces waviness. That is the jointer/naga-dai plane's job [N15, mechanism-by-analogy].
• Heavy stock removal. A thin, lightly-set finishing blade — push it to hog material and it digs and chatters. Bulk removal belongs to the saw, the Drawknife, or roughing planes [N9][N15, mechanism].
• End grain. End-grain work is demanding and highly setup-dependent for any plane; the nankin-kanna's short curved sole and typical bedding make it awkward for sustained end-grain paring compared with a dedicated block plane, but a low bed angle is not itself a disqualifying factor — a low angle can suit end grain well on other tools. Light cuts, a very sharp edge, and supported fibres can make limited end-grain work on a curve possible; treat it as demanding rather than categorically off-limits [N-anat, mechanism-by-analogy].

dangerous (injures the user):

• An exposed blade on a very short sole. The edge projects from a small body and is easy to misjudge by eye; during fettling, sharpening, and setting the tool down it is the exposed thing. Rest it sole-down or sheathed, keep hands clear of the projected travel path, and do not adjust, tap, or clear the mouth while the edge is exposed toward a hand [N10, mechanism; no documented nankin-kanna incident found, labelled].
• Working close to the body, either direction. Pushed or pulled, the tool runs near the operator's hands, forearms, and torso. Secure the work, use light cuts with controlled starts and stops, and keep the body out of the tool's line of travel — general edged-tool precautions rather than a documented nankin-kanna-specific failure mode [N15, mechanism-by-analogy; no nankin-kanna incident found].
• Losing track of the wings on tight work. On a narrow-radius piece the wings ride close to the work and to any clamp; keep the piece consistently held and watch wing clearance through the stroke. Wear eye protection where chips or fragments may be thrown [N15, mechanism].

damaging (ruins the tool or the work):

• Overworking the hard steel when sharpening. The laminated blade is a hard hagane layer forge-welded to a soft jigane body, and the nankin-kanna blade is ground thinner/shallower than a same-size standard kanna [N-anat] — so overheating it on a powered wheel, excessive hollowing, or needless removal of the hardened cutting layer shortens the blade's useful life and can distort its geometry faster than on a standard-size blade. Preserve the back geometry and follow the maker's sharpening guidance rather than grinding by feel [N11][N10, mechanism].
• Wrong sole for the curve — corners gouge. A flat-ish sole forced into a tight concave catches its corners on the hollow's walls before the blade reaches the bottom, gouging instead of fairing; a too-convex sole on a gentle curve rocks and chatters. The sole must suit the class of work [N6, mechanism].
• Tear-out against rising grain on a curve. Curved furniture parts reverse grain mid-piece; planing into rising grain tears out, and on a finished curve the recovery cost is high. Read and follow the grain around the curve [N12, mechanism].
• Forcing a dull blade. A dull edge needs pressure, and pressure breaks the light, balanced two-wing hold the curved cut depends on — producing chatter, facets, or dig-in [N15, mechanism-by-analogy].

beginner-trap (looks fine until it isn't):

• One sole for every curve. A beginner with only the standard slightly-curved sole forces it onto concaves it can't reach and convexes it rocks on, and blames the tool — when the fix is a different (or reshaped) sole. The geometry-to-curve match is the real skill, and it is invisible to the newcomer [N6, mechanism].
• Uneven pressure on the two wings. More force on one wing tilts the tool and cuts deeper on that side; on a round spindle this leaves a flat facet that only shows under raking light — looks fine in the hand, wrong on the bench [N15, mechanism-by-analogy].
• Blade set too deep, mistaken for dullness. A too-far-out blade chatters and ridges; the beginner reads "dull," sharpens, and the chatter stays — because the cause was projection, not edge. Tap-adjustment rewards tiny increments [N11][N15, mechanism].
• Never finding the bite/skew angle. The wrist-rotated contact angle that makes the blade cut does not transfer from flat hira-kanna habit; without it the tool clogs and skates, especially in tight curves, and the beginner plateaus thinking the tool is the problem [N6][N-use].

lineage-misuse (treating it like its sibling, cousin, or twin):

• Like the Sen (sole-less freehand). The cluster's signature pairing and the inverse of the sen entry's own note. The sen has no sole — depth is all hand; the nankin-kanna's sole governs depth once it rides the work [N18][N11]. Lift the wings so the sole floats and only the blade touches, and you have turned a controlled plane into a bad sen — no depth stop, runaway chatter. Conversely, expecting a sen to self-limit like the nankin-kanna makes it dive. They split exactly on the sole [N18][N15].
• Like the flat Kanna (hira-kanna). Apply bench-plane intuition — reference the sole on a flat, take progressive passes to flatten — and the curved sole gives you a curve, not a flat. They are separate categories for a reason [N5].
• Like the Western Spokeshave (push-only, screw-set). A Western woodworker who pushes-only and expects a thumbscrew depth adjuster misreads the tool: the nankin-kanna is tap-set and is used pulled or pushed by grain and feel [N6][N11]. Treating push as the only valid stroke throws away half the tool.
• Like the Drawknife (no depth control, heavy removal). Drive it at drawknife stock-removal rates and the thin, lightly-set blade overloads and the tool chatters and digs rather than removing stock cleanly; treat heavy stock removal as the drawknife's or a roughing plane's job, not this tool's [N15, mechanism-by-analogy].
• Like the Card Scraper (burr hook, high angle). The card scraper cuts with a burnished hook at a steep flex; the nankin-kanna cuts at ~28° with a ground edge. Scraper-style high-angle pressure makes it dig the mouth into the work rather than cut cleanly, and can stress the blade-to-body fit [N15, mechanism].

Correct use (brief — the workbench holds the deep operation)

• The stroke is ambidirectional — and that is itself the tool's signature. Standard dai-kanna are pulled; the Western spokeshave is pushed; the nankin-kanna is documented being used both ways, case by case — a Japanese furniture-maker finds push feels better, a vocational-trained user pulls, and most experienced hands switch by grain direction and stroke length [N6]. Both hands ride the wings; the wrists rotate to find the angle where the blade bites, then hold that angle through the stroke [N6][N-use].
• It fairs curves; it is not built for dimensioning flats. Its genuine strength is finishing convex, concave, and round work — chair parts, spindles, round bars, curved edges — to a fair surface, often cleaning up after a router or saw [N6][N9]. It is primarily a controlled shaping and fairing tool — most examples are set up for light-to-moderate cuts rather than drawknife-style stock removal, though cut aggressiveness depends on blade set, mouth opening, sole, and the wood itself; Okubo Kotaro's high-volume production use shows the class is not confined to fine finishing alone [N6][N9][N8].
• Choose a sole suited to the curve, and read the grain. The sole's curvature should broadly suit the work (flatter or gently relieved for convex, more strongly convex for concave) — not necessarily an exact radius match [N6]; plane with the grain, expecting grain reversals mid-curve on furniture parts [N12]. A slight skew to the stroke clears shavings and eases the cut [N-use].
• Let the sole govern depth. Unlike the sen, the sole self-limits the cut once it rides the work — keep the sole in contact, take light passes, "like a plane landing and taking off" to avoid start/stop lines [N12]. (Deep operation, sole-fitting, and sharpening sequence are Reed's — see the handoff.)

Matching

• does: Fairs and finishes convex, concave, and round wood surfaces that a flat-soled plane cannot reference — chair legs, arms, backs, and curved seat/panel edges; spindles, round bars, and mouldings; spoon, spatula, and treen blanks; and, in the dictionary's narrower traditional sense, circular grooves [N1][N6][N9].
• suits: Furniture-makers and chair-makers fairing curved parts; spoon/utensil/treen carvers working at volume; general carpenters cleaning up round stock after a router or saw; makers already comfortable reading grain by feel and switching push/pull stroke by feel rather than by fixed rule [N6][N8][N12].
• kit-fit: Pairs with the flat Kanna (dai-kanna) for flat work and the Sen for bespoke, freehand curved fitting; sits alongside the convex-soled sori-kanna and the small-curve hokkecchi-kanna sub-variant as sole-geometry options within the same family; bought semi-finished like other Japanese planes, so it arrives as a fettling project, not a ready-to-use tool, unless bought from a maker (like the historical Azuma Atushi) who finished them ready-to-use [N5][N7][N10].
• not-for (project-keyed):
  • Flat panel/face truing and thicknessing → use the Kanna (hira-kanna).
  • Long straight jointed edges → use a jointer/naga-dai plane.
  • Heavy stock removal / rough shaping → use the Drawknife or a roughing plane.
  • Sustained end-grain work → possible only with care (light cuts, sharp edge, supported fibres); a dedicated block plane is the better-suited tool.
• access: Sold new, semi-finished, by specialist Japanese toolmakers/retailers — Tsunesaburo's SHUNRAN line via Fine-Tools, Kurashige (Yoshitaka line), Yamamoto/Suzuki-ya — in widths from 15–48 mm [N10][N11][N14]; blades from named smiths such as Komori Hideki (Echigo Yoita tradition) are also available [N9]. Buyers should expect to rework the mouth, wings, and sole before first use. Vintage examples circulate on secondary markets (reaching back only decades, unauthenticated for age) [N-vint].
• revival: Not a revival — a living tradition. Multiple named makers (Komori Hideki, Tsunesaburo/SHUNRAN, Yoshitaka, Yamamoto/Suzuki-ya) are in current production, and named living users (Okubo Kotaro, Big Sand Woodworking's Jon) document active daily use [N7][N8][N9][N10][N12][N13].

Anatomy

Only where the nankin-kanna differs from the family. The laminated-blade construction (Kanna/Japanese chisel) and the spokeshave biomechanics (Spokeshave) are the relatives'; the nankin-kanna's signature is that it is a sole brought to curved work — a short, curvature-matched sole spanning two wings.

• The body (台 dai). A short white-oak (白樫 shira-kashi) block — representative size ~260 × 50 × 24 mm for a 36 mm tool — mortised for a tapered blade, with a wing/handle projecting from each side [N10][N11][N14]. These are the Tsunesaburo/Kurashige/Yamamoto product specs found this run; treated as representative examples of current commercial practice, not as fixed universal anatomy — dimensions, bedding, and materials vary by maker and by the radius the tool is built for. Sold as a semi-finished blank: the owner reworks the mouth, shapes the wings to the hand, and trues the sole before use — a day or two's fettling, normal for Japanese professional planes [N7][N10].
• The blade (刃) — laminated, thin, no chipbreaker. Standard Japanese awase lamination: a hard hagane edge (white steel 白紙 / blue steel 青紙) forge-welded to a soft jigane body (極軟鉄) [N11][N10]. Two anatomical specifics matter: it is bedded flatter than a bench plane — ~25–28° (lower than the dai-kanna's ~38°) [N11], and it is shaped thinner and shallower than a same-size standard kanna blade [N-anat]; and traditionally it carries no chipbreaker [N10]. (The 25° vs 28° figures both appear across makers; held as maker-variation, not collapsed.)
• The sole (台裏 dai-ura) — the whole identity. Slightly curved, and shaped or chosen for the class and radius of the work [N10][N6] — not necessarily an exact geometric match; sole length, bearing area, blade projection, and mouth all factor into whether a given sole reaches a given curve. This is the structural heart of the entry: where the dai-kanna's sole is dead-flat to ride a board, the nankin-kanna's is curved to ride a round. The toolmakers describe the working rule — "by varying the curvature of the sole, it finishes almost any concave or convex curve" [N6] — and practitioners treat it as a matter of degree, not precision-fit: a more strongly convex sole reaches tighter concave/hollow work (the sori-kanna principle), a flatter or gently relieved sole suits broader convex work like spindles and round bars [N5][N15, mechanism]. A brass or metal wear-insert at the mouth is common [N5].
• The mouth (刃口 ha-guchi). Present — it is a bodied plane — and is the part the owner most reworks on the blank to set chip clearance [N10].
• Depth adjustment — tap-set, no screw. Like every traditional dai-kanna: the tapered blade is its own wedge; tap the blade head to advance, tap the body head to retract [N5][N11]. No screw mechanism (the point of divergence from the metal Western spokeshave, many of which adjust by thumbscrew).
• The two wings/handles. One per hand, of the same oak, deliberately left "bulky and angular" so the owner can carve them to fit [N10]. The two-hand-on-the-wings hold is the tool's defining grip.
• What it HAS that its siblings do not: a sole, a mouth, a body governing depth — exactly the Sen's and Yariganna's absence list, inverted. That inversion is the cluster's whole point [N11][N18].

Variations (and what it is NOT)

• By sole geometry (the meaningful axis). A near-flat / slightly-curved sole for convex work — rounding spindles, round bars, chair legs (the standard production form: SHUNRAN, Yoshitaka, Yamamoto) [N10][N11][N14]; a convex sole for concave/hollow work [N6][N5]. The principle (match sole curvature to the class of work) is firmly sourced; a dedicated convex-sole nankin-kanna as a current retail SKU was not confirmed — practitioners often reshape the sole of a standard tool, or own several [N6][N12]. Held: principle CONFIRMED, distinct-SKU NOT confirmed; not necessarily an exact radius match (see Anatomy).
• By size. Blade widths in current commerce run 15, 18, 24, 30, 36, 42, 48 mm, each on a correspondingly-sized body [N14][N10][N11].
• By blade steel. White steel (白紙) and blue steel (青紙) laminated blades; some modern lines in HSS (SKH51) [N10][N11].
• By region. Two confirmed production centres: Echigo Yoita (Niigata) — Komori's small-plane-blade tradition [N9][N11] — and Banshu Miki (Hyogo) — Tsunesaburo's SHUNRAN [N10].
• Named sub-variant — hokkecchi-kanna (剥抉鉋). A nankin-kanna with a different blade-fixing method, for smaller curves [N5].
• By body material. Wooden (white-oak) body is the form found in every current-production example located this run; a brass/metal-bodied nankin-kanna was NOT found in current production (the Western Spokeshave has metal bodies) — but absence-in-search is not the same as absence-in-practice, so this is held as "not found," not "does not exist" [gap].
• What it is NOT: not a flat-truing plane (that's the Kanna/hira-kanna — its curved sole cannot reference a flat); not a sole-less freehand scraper (that's the Sen — the nankin-kanna's sole governs depth); not a low-skew paring spear blade (that's the Yariganna); not a burr-hook plate (that's the Card Scraper); not a no-depth-control draw blade (that's the Drawknife). Each line is drawn in How NOT to use it.

Trades (where it appears, with trade-specific notes)

• Furniture-making (家具職人) — the tool's home. Fairing curved furniture parts; "furniture makers cannot do without it" [N6]. Includes the Matsumoto Mingei Kagu folk-furniture world (Azuma) [N7].
• Chair-making. Arms, backs, legs, curved seat edges — the application Big Sand Woodworking and the Sawmill Creek discussions center on [N12].
• Spoon / utensil / treen carving. Okubo Kotaro's whole practice; round and hollowed kitchenware at production volume [N8][N13].
• General carpentry / round-stock finishing (大工). Rounding round bars, spindles, mouldings; cleaning up after router or saw [N9][N-use].
• Honest edge: documented use in instrument-making (shamisen/koto/lutherie), comb-making, and fan-making is claimed in a single retailer line but NOT corroborated by a named maker this run — plausible by function, not asserted [gap]. Outside Japanese (and Chinese-derived) practice its work is done by the Western Spokeshave.

Lineage

The "what a curved-work surfacing tool is" history belongs to the Sen / Yariganna / Spokeshave parents and is not re-derived here. The nankin-kanna's lineage is the bodied-plane clock laid out in the Kanna entry, crossed with the surfacing-tool family the Sen and Yariganna entries built. Note throughout: the comparisons below are a structural framework Hewn draws across the cluster, not a documented sequence of invention — no cited artifact series, text, or maker testimony shows one tool giving rise to the next.

• Ancestors — the body, and the blades that had none. The nankin-kanna is a member of the bodied-plane family: it shares the dai-kanna's blade-bedding (a tapered laminated blade that is its own wedge in a mortise, no separate wedge) [N5], and that whole rectangular bodied form entered Japan from China in the mid-Muromachi period (~15th c.) [N4]. Its closest functional analogue on the curved-work side is the sole-less pulled blade — the Sen draw-scraper and, behind it, the Yariganna — tools that answer the same curved-work problem without a body. The nankin-kanna is, structurally, what results if you give the sen's job a small curved sole — a comparison Hewn draws for clarity, not a documented sequence of invention.
• Contemporary — the Sen (the load-bearing sibling comparison). This is the spine the Sen entry already drew: the sen was not displaced by the bodied plane, because round work has no flat to reference; the nankin-kanna is the bodied tool that does reach round work, by curving its sole. They are separately-categorized, both-current tools — the toolmaker Kurashige lists 南京鉋 and 銑 (sen) as distinct product lines [N11] — and neither displaced the other. The reason is a real division of labour: the nankin-kanna gives a repeatable, sole-governed profile (the sole self-limits depth); the sen gives a bespoke, freehand fit (the cooper's lightless joint, no sole, all hand). Same curved-work territory, two different promises.
• Sibling — the flat Kanna (dai-kanna). The nankin-kanna is the curved-work member of the bodied-plane family whose flat-work member is the dai-kanna. Same blade system, opposite sole: the dai-kanna's sole is dead flat to true a board; the nankin-kanna's is curved to fair a round. (The Japanese taxonomy files it under sori-dai-kanna, curved-sole planes [N1].)
• Cross-tradition twin — the Western Spokeshave. The clearest convergence in the cluster: a small two-handled bodied curved-work plane arrived at independently in Europe and, in Japan, on a bodied-plane form generally understood to trace to China [N4] — though this specific tool's route of transmission is unverified — and English-language sellers simply translate one as the other [N10][N11]. The honest difference is stroke and body: the Western spokeshave is pushed and typically metal-bodied; the nankin-kanna is wood-bodied and ambidirectional in practice (see Use). A near-relative on the Japanese side, the convex-soled sori-kanna, scoops concave surfaces [N5]; a named sub-variant, the hokkecchi-kanna (剥抉鉋), is a nankin-kanna with a different blade-fixing for smaller curves [N5].

Petroski lens — the failure a curved sole answers, held as a structural reading, not a historical one. Across this cluster "form follows failure" has run as a stack of fixes culminating in the body — but the Sen entry showed the ratchet stopping at round work, because a flat sole has no purchase on a curve. Read structurally, the nankin-kanna answers that failure: not "discard the body" but "curve and shorten the sole until it rides the round." The short, slightly-curved (and curvature-matched) sole is the flat sole reinvented for work that has no flat — depth governed by the sole again, but on a surface the dai-kanna could never touch. Hewn does not read this as a documented lineage. The tool wears a name reported as a foreign-marking label applied by tool merchants from the Meiji period [N7][N2]; the bodied form it belongs to is understood to be a Chinese-derived import [N4]; and no pre-Meiji Japanese record of this specific tool was found [gap]. Those facts leave the tool's date and route of arrival open, not settled — the record does not establish that the curved-sole fix was evolved in place from the yariganna→kanna line, but it equally does not establish that it "arrived sideways" as a late import; either is possible, and Hewn holds the question open rather than resolving it into a historical claim. (Held as the entry's genuinely-new structural reading — a labelled interpretation built on sourced facts: the curved/short sole [N10][N6], the bodied form's Muromachi-era continental origin [N4], the Meiji-era trade-name [N7], the both-current separate categories [N11], and the sole-governs-depth distinction from the sen [N18][N15]. Distinct from the facts, explicitly a structural comparison across the Run-146/147/148 cluster rather than a proven historical sequence.)

History

Family history lives in Kanna / Sen / Yariganna. Nankin-kanna-specific — and this is the thinnest historical record in the cluster, honestly, and its date is not established either way:

• The dating verdict — NOT ESTABLISHED. No dated artifact, dated depiction, or museum accession for the nankin-kanna was found this run. What can be bounded: the bodied rectangular plane it belongs to arrived in Japan from China in the mid-Muromachi period (~15th c.) [N4], so the bodied form the nankin-kanna belongs to cannot be older than that; the chipbreaker (which the nankin-kanna traditionally lacks [N10]) was a Meiji import [N4][N5]; and the name "nankin" is reported by the expert Kurayuki Abe to have been applied to this tool by tool merchants "since the Meiji period" (post-1868) [N7]. Put together, what the record actually supports is that the named tool cannot be dated earlier than the Meiji period with current evidence — it does not support a positive claim that the tool itself is young. A name can attach late to an older form, and the absence of a pre-Meiji record from one research pass is not proof the tool didn't exist before it was so named. Held honestly: the date is open, bounded on one side by the Meiji naming and the Muromachi arrival of the bodied form, unbounded on the other. Routes to firmer dating: a Takenaka 『収蔵品目録 第4号 鉋篇』 entry, an Edo tool-encyclopedia depiction, or a dated artifact [gap].
• A foreign-marked name; the form's origin less certain. The one thing the history does say clearly is that the tool was labeled as foreign/Chinese-associated — the whole point of the 南京 prefix [N2][N3]. A practitioner reads something similar off its form: unlike the hira-kanna, which "underwent Japan-specific development," the nankin-kanna "has retained much of its original [continental] form" [N6] — though this is one practitioner's impression, not a documented transmission history. Whether the tool physically came from Nanjing/China, arrived as this exact form, or was merely styled as Chinese remains unproven [N16].
• The 20th-century record is where it gets vivid. The tool's best-documented history is recent and human (see Human side): a named Matsumoto furniture-maker, Azuma Atushi (1945–1996), made and sold ready-to-use nankin-kanna at Craft Fair Matsumoto from 1993, and they were exported to the College of the Redwoods fine-woodworking program in California (1993–96) [N7]; and a named living maker, Okubo Kotaro, has built a whole practice around it [N8]. The nankin-kanna's documented life is, unusually for this corpus, mostly within living memory — though that reflects where documentation happens to survive at least as much as it reflects the tool's actual age.
• Where it sits against the sen. The same curved-work niche the Sen has held "unchanged since medieval times" [N-sen], the nankin-kanna entered with a bodied answer — and the two have coexisted in separate, current production ever since [N11]. The history here is not a replacement story (as the dai-kanna was for the yariganna) but a niche-sharing one: a bodied tool, whose arrival relative to the sen is not precisely dated, that took the repeatable-profile half of the curved-work job and left the bespoke-fit half to the sole-less sen.

✦ Human side

The nankin-kanna's people are furniture-makers, chair-makers, and spoon-carvers — and unusually for this corpus, the best human texture is from named, living, photographed hands.

• The furniture-maker — the tool's home trade. A professional Japanese furniture-maker (pen-name "artisan," blogging since 2005) calls the nankin-kanna "a plane that furniture makers cannot do without," prized because "by varying the curvature of the sole, it delivers a quality finish on almost any concave or convex curve" [N6]. This is the tool's center of gravity: fairing the curved parts of furniture — chair legs, arms, backs, curved edges of raised panels.
• Okubo Kotaro (大久保公太郎) — the nankin-kanna's most-documented living hand. A spoon and kitchen-spatula maker in Matsumoto, Nagano, who works ~350 days a year and makes roughly 6,000 pieces annually, almost entirely with nankin-ganna of his own specification [N8][N13]. He soaks wood blanks in water so they "can be shaved like butter" — a technique he found that "past woodworkers, like from the Edo period, often soaked dry wood in water to soften it" — and has fine-tuned blade angle and mouth with blacksmiths over years, keeping every old shave with its settings recorded on the body [N8]. He taught a 2023 workshop in Ashland, Wisconsin [N13]. His stated ethic is its own kind of texture: "I don't have trade secrets. I share everything I learned over the years with anyone" [N8].
• Azuma Atushi (東 敦史, 1945–1996) — the maker who sold it ready-to-use. A craftsman with 26 years at Matsumoto Mingei Kagu (the famous folk-art furniture house) who, from 1993, hand-finished nankin-kanna and sold them at Craft Fair Matsumoto — "twenty-odd pieces, and when one person picked up several at a time they were gone" — unusual in Japan, where nankin-kanna are normally sold as semi-finished blanks the owner must fettle. He exported ~30 to the College of the Redwoods program in California before his death at 50, in January 1996 [N7]. Documented by the joinery expert Kurayuki Abe, who also preserves the print record (MANO No. 2, 1996) [N7].
• The named blade-smith — Komori Hideki (小森英樹, b. 1952). A third-generation small-plane blade smith in the Echigo Yoita (Niigata) tradition who makes nankin-kanna blades specifically; certified a 伝統工芸士 (Traditional Craftsman) in 2001 [N9]. His heritage page describes the tool plainly: "Unlike ordinary planes, [nankin] planes are equipped with plane blades along the grain of the wood. It is mainly used to cut curved parts such as furniture" [N9]. A named, officially-recognized maker is the firmest human anchor the entry has.
• The Western hand who made his own — first-person texture. Jon of Big Sand Woodworking (a Japan-based furniture-maker) forged a nankin-ganna for a chair and wrote the cluster's most vivid use-quote: "This tool makes shaving wood on curved surfaces ridiculously easy… much easier on your hands to hold for long periods… The nankin-ganna almost has to hit the surface like a plane landing and taking off, otherwise you end up with really distinct lines where the plane starts and stops" [N12].
• The push-or-pull debate — a living cultural seam. The nankin-kanna sits exactly on the fault line between the Japanese pull tradition and the Chinese/Western push: the furniture-maker "artisan" confesses "unlike other dai-kanna where I pull, with the nankin-kanna I find pushing has a better feel — why is that?" and his commenters (including a vocational-school-trained pull-user) confirm they use both, case by case, depending on grain and the length of the cut [N6]. The tool's foreign origin shows up in the hands, not just the name.
• Honest pre-modern gap. Every named hand above is 20th–21st century. No named pre-modern maker or user of the nankin-kanna was found — consistent with the thin pre-modern record generally, though this reflects an absence of surviving named accounts rather than positive evidence of the tool's age [N7][gap].

✦ Proof of existence

The nankin-kanna's proof profile rhymes with the Sen's — strong present-tense, thin past-tense — but for a different reason: the sen's thin antiquity record sits on a tool independently dated old by other means; the nankin-kanna's thin antiquity record sits on a tool whose age is simply not established — it could be old with a late name, or genuinely recent; the evidence does not decide between them.

• Living production (strong, present-tense proof). Multiple named makers in current production, sold and photographed by established toolmakers: Tsunesaburo's SHUNRAN line (Banshu Miki; white-oak body, Aogami #1 / SKH51; 30–48 mm) via Fine-Tools and Kurashige [N10][N11]; Komori Hideki's small-plane blades (Echigo Yoita; a 伝統工芸士) [N9]; Yoshitaka / Yamamoto / Kouetsu lines, 15–48 mm [N11][N14]. Existence-in-use today is unambiguous and arguably better-anchored in named officially-recognized makers than the sen [N9][N10].
• Documented living use (strong, via named hands). Okubo Kotaro — ~6,000 pieces a year, photographed in his studio (Entoten, Woodspirit Handcraft) [N8][N13]; Azuma Atushi — made, sold, and exported the tool, documented by the joinery expert Kurayuki Abe with a print citation (MANO No. 2, 1996) [N7]; Big Sand Woodworking — first-person chair build with photographs [N12]. The use record is firmer and more recent than most of the corpus.
• Lexical proof. Entered in a standard Japanese dictionary (精選版 日本国語大辞典) as a defined headword [N1] — firmer than the Sen, whose 銑 carries no general-dictionary tool sense.
• The antiquity / depth record — the weakest in the cluster, and genuinely open. No excavated or dated specimen; no rendered museum accession (the Takenaka Carpentry Tools Museum holds a plane collection and a published 鉋篇 catalogue, but no nankin-kanna holding was confirmed online this run); no pre-Meiji depiction or document; and the name itself is reported as a Meiji-era tool-merchant label [N7][N4]. Vintage secondary-market examples reach back only decades [N-vint]. Held honestly: this is the weakest depth record in the cluster, and the absence of pre-Meiji evidence supports an open dating verdict — it does not, by itself, establish that the tool is young; it only establishes that its documented history does not currently reach that far back.
• Route to firmer grounding: a Takenaka 『収蔵品目録 第4号 鉋篇』 / 文化遺産オンライン nankin-kanna accession; an Edo or Meiji tool-encyclopedia depiction that would date the named tool; or Toshio Odate's actual passage read directly [N16]. These would either confirm an Edo presence or firm up the "Meiji-named, recent" reading — or, equally, rule it out.

references

```
[N1] Kotobank — 精選版 日本国語大辞典 (Seisen Nihon Kokugo Daijiten), entry 南京鉋: reading
なんきんかんな; def. "反り台鉋の一つ。円溝を削るのに用いる小形のもの" (a type of sori-dai-kanna;
a small one used for cutting circular grooves). https://kotobank.jp/word/南京鉋-348454
— kind: dictionary (HIGH for the name + a defined sense; note the def names a NARROWER job
[circular grooves] than the broad curve-finishing practitioners describe — flagged). research passes-sourced.
[N2] National Diet Library reference database (crd.ndl.go.jp), response 2010-01-11, citing
『暮らしのことば 新語源辞典』(Yamaguchi Yoshinori ed., Kodansha, 2008): the 南京 prefix as a
foreign-origin marker that displaced older 唐 (kara) in the early-modern period — 南京豆 (peanut),
南京虫 (bedbug), 南京更紗 (Nanjing cotton), 南京錠 (padlock, "a lock of foreign origin").
https://crd.ndl.go.jp/reference/entry/index.php?page=ref_view&id=1000087410 — kind: national-
library reference citing a published lexicon (HIGH for the prefix etymology). research passes-sourced.
[N3] Kotobank — デジタル大辞泉 (Shogakukan), entry 南京 as 造語要素: ① "中国産のもの、または中国方面
から渡来したものの意" ② "珍しいもの、または小さくかわいらしいものの意." — kind: dictionary
(HIGH for the two prefix senses). research passes-sourced.
[N4] Takenaka Carpentry Tools Museum (竹中大工道具館) — dougukan.jp/tools/24: the rectangular bodied
plane (台鉋) introduced to Japan ~mid-Muromachi from China; the double-blade/chipbreaker (二枚刃)
a Meiji import from Western planes. https://dougukan.jp/tools/24 — kind: museum (HIGH for the
bodied-form date + the chipbreaker date; does NOT name the nankin-kanna specifically). research passes-sourced.
[N5] Wikipedia — "Japanese plane" (EN) / 鉋 (JA): nankin-kanna = "a spokeshave with two handles";
sori-kanna (反り鉋) = convex-soled plane for scooping concave surfaces; hokkecchi-kanna (剥抉鉋)
a nankin-kanna sub-variant with a different blade-fixing for smaller curves; tapered blade is its
own wedge (no separate wedge). https://en.wikipedia.org/wiki/Japanese_plane — kind: encyclopedia
(MEDIUM). research passes-sourced.
[N6] Koubou Tsushin Yuyuu — "南京鉋は、Push or Pull?" (koubou-yuh.com/blog/?p=140, 2010), professional
furniture-maker's blog (pen-name "artisan," active since 2005) + 6 named-commenter thread.
"曲面切削で活躍するのが南京鉋… 家具職人としては欠かすことのできない鉋"; vary the sole's curvature for
concave→convex; push-vs-pull genuinely contested (author prefers push; vocational-trained commenter
pulls; most use both); the name "clearly" marks continental origin; the tool "retained much of its
original form." — kind: established practitioner blog (MEDIUM). research passes-sourced.
[N7] Kurayuki Abe (木とジョイントの専門家) — abeshoten.jp/blog/2951: Azuma Atushi (東 敦史, 1945–1996,
26 yrs at Matsumoto Mingei Kagu) hand-finished, ready-to-use nankin-kanna sold at Craft Fair
Matsumoto from 1993 (sold out instantly), ~30 exported to College of the Redwoods (Fort Bragg, CA),
d. Jan 1996; print record MANO No. 2, 1996, pp.18–21. Key dating claim: the name "南京" was applied
to this tool by 道具商 (tool merchants) "since the Meiji period." Abe proposes dropping the
"meaningless" 南京 for 棒鉋 (bō-kanna). http://kurayuki.abeshoten.jp/blog/2951 — kind: named expert
with archival access + print citation (MEDIUM-HIGH). research passes-sourced.
[N8] Entoten — Ai Kanazawa, "Eat, Sleep, Shave Wood: The Extraordinary Life of Woodworker Okubo Kotaro"
(2023; studio visit, photos by Douglas Matsumoto). Okubo (Matsumoto, Nagano): ~6,000 spoons/spatulas
a year, ~350 days, almost entirely with self-spec'd nankin-ganna; soaks blanks in water (an Edo
technique); co-develops blades with smiths; keeps every old shave with dated settings; "I don't have
trade secrets." https://www.entoten.com/2023/11/24/eat-sleep-shave-wood-the-extraordinary-life-of-woodworker-okubo-kotaro/
— kind: named journalist, named subject, studio photos (HIGH for living use). research passes-sourced.
[N9] Yoita-Uchihamono heritage site (yoita-uchihamono.com) — Komori Small Plane Factory: Komori Hideki
(小森英樹, b.1952), 3rd-gen small-plane blade smith, certified 伝統工芸士 (Traditional Craftsman) 2001;
makes nankin-kanna blades — "Unlike ordinary planes, [nankin] planes are equipped with plane blades
along the grain of the wood. It is mainly used to cut curved parts such as furniture."
https://yoita-uchihamono.com/global/craftsman/komori/ — kind: official craft-heritage site
(HIGH for the named, recognized maker). research passes-sourced.
[N10] Fine-Tools / Dieter Schmid — SHUNRAN Spokeshave (Nankin Kanna) by TSUNESABURO (Miki City, Hyogo):
white-oak body, Aogami No.1 / SKH51 blade, no chipbreaker, "the sole is slightly curved," body
~50×260×24 mm (36 mm tool), widths 30/36/42/48 mm, sold semi-finished (user reworks the mouth).
https://www.fine-tools.com/SHUNRAN-Spokeshave-Nankin-Kanna-Aogami-No.1-Steel-Blade-Width-36-mm-TSUNESABURO/323946
— kind: established commercial maker/retailer (HIGH for living production + anatomy specs). research passes-sourced.
[N11] YAMASUKE Kurashige Tools (Tokyo, est. 1959) — 南京鉋 (Spokeshave Plane) collection + product pages:
lists 南京鉋 and 銑 (sen) as SEPARATE categories; Yoshitaka line ~28° bedding, jigane 極軟鉄, Echigo
Yoita; blue/white steel; tap-adjusted; widths 24–42 mm.
https://shop.kurashige-tools.com/en-us/collections/spokeshave-plane-nankin-kanna — kind: specialist
retailer (HIGH for living production + the sen/nankin-kanna category split + bedding angle). research passes-sourced.
[N12] Big Sand Woodworking — Jon (Japan-based furniture-maker), "Chair No.1 Fabrication Part 2": self-made
nankin-ganna for chair arms/backs; "shaving wood on curved surfaces ridiculously easy… easier on your
hands… has to hit the surface like a plane landing and taking off, otherwise you get distinct
start/stop lines"; managed a clean cut against rising grain at a joint.
https://www.bigsandwoodworking.com/chair-no-1-fabrication-part-2/ — kind: named woodworking blog,
photos (MEDIUM-HIGH for first-hand use). research passes-sourced.
[N13] Woodspirit Handcraft — Jarrod Dahl, "Kotaro's Workshop" (2023): Okubo "carves 360 days in the year
and makes 6,000 each year" with the nankin kanna ("he has many shapes and sizes"); taught a workshop
in Ashland, WI. https://www.woodspirithandcraft.com/blog/kotaros-workshop — kind: named teacher/host
blog (MEDIUM-HIGH). research passes-sourced.
[N14] Suzuki-ya / Yamamoto — nankin-ganna product page: white-oak body, laminated blade, widths 15–36 mm.
suzukitool.com (Yamamoto spokeshave plane / nankin-ganna) — kind: commercial retailer (HIGH for
production + size range). research passes-sourced.
[N15] Wikipedia "Spokeshave" + Taylor Tools / Fine Woodworking forum — the sole self-limits depth (the
spokeshave-vs-drawknife distinction); concave-bottom shave for outside/convex curves; chatter from
too-deep blade or dullness; jointing/long-edge limits of a short sole. Used for the lineage-misuse and
mechanism-by-analogy items. https://en.wikipedia.org/wiki/Spokeshave — kind: encyclopedia/retailer/
forum (MEDIUM / LOW-MEDIUM; mechanism-by-analogy, labelled). research passes-sourced.
[N16] Toshio Odate, "Japanese Woodworking Tools: Their Tradition, Spirit and Use" (Taunton, 1984) — relayed
via Sawmill Creek (2016): the planes are "probably named after a city in China, Nankin, where they
probably originated" (hedged twice). Cited as authority; the actual passage NOT read directly this
pass — flagged. — kind: book via forum relay (MEDIUM as authority; not directly verified).
[N17] Internal corpus — Sen (Haft Run 147), Yariganna (Run 146), Kanna (Run 085): the Sen
entry names the nankin-kanna as "the closest thing to a partial successor for repeatable curved
profiles… a separate, still-current category alongside the still-current sen," and as the "sole-
registered curved plane" the sen is NOT. This entry builds out that named sibling and closes the
surfacing-tool cluster. Family parents/cousins: Spokeshave, Drawknife, Card Scraper,
Adze, Travisher, Shaving Horse. — kind: internal corpus.
[N18] BladeForums — Japanese sen / nankin-kanna distinction: the sen "would not have a sole, whereas the
nankin kanna, being a spokeshave, would have a sole for better control." — kind: forum (MEDIUM for the
explicit sole distinction, corroborated by [N11]'s category split). research passes-sourced.
[N-anat] Japanese-language retail/usage notes (daiku-dougu.jp et al.) — the nankin-kanna blade is "刃の丈と
厚みが薄く" (shaped thinner/shallower than a same-size standard kanna blade); wrist-rotated contact
angle found then held; HSS blades ground wet to avoid burning. — kind: retailer usage guide (MEDIUM;
direct fetch had encoding issues, confirmed from search excerpts — flagged). research passes-sourced.
[N-use] daiku-dougu.jp / Sawmill Creek — usage: rotate the wrist to find where the blade contacts, then
maintain that angle; a slight skew clears shavings. — kind: retailer guide + forum (MEDIUM / LOW).
research passes-sourced.
[N-vint] WorthPoint secondary-market archive — multiple sold "Vintage Japanese Nankin-Kanna / Spokeshave"
listings (24/36 mm), one described ~40 years old (c.1974). — kind: secondary market (LOW-MEDIUM;
seller-labelled, unauthenticated; reaches back only decades). research passes-sourced.
[N-sen] Internal corpus — Sen (Run 147): the sen "virtually unchanged in form since medieval times,"
held in active production; the curved-work niche the nankin-kanna entered later. — kind: internal corpus.
```

The reviewer's corrections (14)

• BLOCKER — 1. Correct Komori Hideki's credential status
• BLOCKER — 2. Recast the origin and transmission history as unresolved
• SHOULD-FIX — 3. Stop using the name itself as proof of foreign origin
• BLOCKER — 4. Correct the Western-spokeshave comparison
• SHOULD-FIX — 5. Add the dated documentary record and narrow the historical gap
• SHOULD-FIX — 6. Correct the museum-evidence statement
• SHOULD-FIX — 7. Make dimensions and blade angle model-specific
• SHOULD-FIX — 8. Treat anatomy as a variable family, not a universal form
• SHOULD-FIX — 9. Qualify the setup-time claim
• SHOULD-FIX — 10. Moderate the mechanics claims
• SHOULD-FIX — 11. Remove absolute task exclusions
• SHOULD-FIX — 12. Replace inferred safety prose with sourced controls
• SHOULD-FIX — 13. Recalibrate confidence by claim class
• MINOR — 14. Remove workflow residue from the public entry