Tips for collecting Nihonto

Note: This rather long but useful excerpt was written by a well-known collector who would prefer to remain anonymous.

A topic that comes up regularly is the question "is this sword worth buying?" or "is this signature genuine?", usually followed by some blurry pictures of a rust-covered blade in battered Guntô mounts. The answer is pretty simple: it's worth every penny you're willing and able to pay for if all you're looking for is a weapon of Japanese origin. Might there be a collectible sword under all that surface erosion, or is it even a wise investment to buy a sword like that? In my opinion, chances are as high as finding a pot of gold at the end of the rainbow.

It usually takes decades of experience - and preferably a traditionally-trained togishi (polisher) - to judge whether a corroded blade might be restorable or not, and even then I've seen the expression of an unpleasant surprise crossing the faces of those experts after Madogaki (polishing a "window" for evaluation). And this happened after hands-on examination by extremely knowledgeable persons. I therefore strongly advise against buying such a sword. Especially the beginning collector fares much better with a Nihontô in a reasonable state of polish.

But first of all, we have to ask ourselves the question why we want to buy and/or study Japanese swords; if the answer is "because I just want a Japanese sword - any Japanese sword, and am not willing to pay more than $$$," read no further. If you're fascinated by Nihontô as an art form, if you're "bitten by the bug," and prepared to explore a whole new world of marvels in steel and interesting Oriental history, you may find some of my following thoughts at least mildly entertaining.

What should we collect and how should we collect? In putting together my line of reasoning (as I like to think of it), I wasn't too surprised to learn that already others had tackled that problem before me. Being basically a lazy-bone, I borrowed heavily from Raymond Bushell's chapter on new principles of collecting in his book "Netsuke - Familiar & Unfamiliar." I shamelessly adopted and re-wrote much of what was said there, in part even descending to the netherworld of outright plagiarism. However, I am solely responsible for the content of this post - it is, after all, my point of view on Nihontô, and not necessarily that of the author whose writings I (ab)used.

First I should explain what exactly I'm going to talk about. Maybe the translation of the name of the NBTHK can help: Nippon Bijutsu Tôken Hozon Kyôkai means not only society for the preservation of the Japanese "sword," but also of the "art sword," clearly stating that the goal is not only the preservation of the sword as a weapon but also as an art object.

But what is art, and what kind of Nihontô qualify as Bijutsu Tôken? Beauty, of course, lies in the eye of the beholder, and even "the experts" are not always in agreement. However, borderline cases are few and far between as certain standards and "laws" have been established and are universally acknowledged.

The artistic features of the Japanese art sword can be recognized and studied as they can be shown and explained. This knowledge has nothing to do with spiritual studies, Zen-Buddhism, enlightenment, Iaidô training, or sentiment; it is a question of mere study. The same methods applied to recognize architecture, paintings, sculptures, and music according to their styles can be used for swords, which can be dated and allotted in a school, province, etc.

The ultimate preparation available to the collector who would like to find art swords is at once the most elementary and the most sophisticated preparation of all. It is to learn the subject. No one knows instinctively what a good Nihontô looks like, nor does anyone know intuitively the elements that constitute it. The collector must absorb the basics in a gradual accretion of understanding. Likewise, good taste in Nihontô is not an instantaneous revelation. It's a gradual development. Most collectors readily concede the improvement of their tastes over previous years. Good taste requires careful nurturing and tending for a mature blossoming. A natural good eye means a head start, an enviable beginning, but it is not enough. Just as a good voice without musical training will not enable one to sing like Pavarotti, so also a good eye will not assure a fine collection without some application and study.

In order to appreciate the different types of beauty, one should be equipped with as much knowledge as possible and a seeing eye regarding a good blade. Therefore it is useful to memorize the characteristics of the different "roads," schools, and masters, so that when looking at a Nihontô one knows where, when, and by whom it might have been made. This is the only basis on which to achieve judgement about the differences in quality.

The collector who boasts "I don't know anything about Nihontô; I just buy what I like," makes a statement that is not very profound. Of course he buys what he likes. If he doesn't buy what he likes, what does he buy? If he doesn't buy what he likes, he had better not collect. The collector who doesn't know anything about Nihontô will benefit by learning. If he should be blessed with innate good taste, he may develop expertness by listening and looking, like gifted students who earn degrees without cracking a book. For most of us, however, reading, discussing, examining, and studying comprise an essential, though happy, regimen for graduation to connoisseurship. The emotional response to a superb Nihontô may be as intense for the collector who never learned any "technical" information as for the expert, just as the emotional response to music may be as great for the listener who can't hum a tune as for the trained musician. But the intellectual pleasure, if not the emotional response, of the musician is profoundly enhanced by his understanding of theme, harmony, and counterpoint. So also is the intellectual pleasure greater for the collector who understands sugata, hada, hamon, hataraki, school, smith, etc.

It also needs to be mentioned that the features found in a good blade become only obvious and recognizable when brought to light by a skillful polisher. The togishi needs to know how the blade was originally designed in order to bring out the characteristics the smith intended, as each style requires a slightly different technique. A poor polisher can change the character of a blade in a way that a good old Kotô blade looks like an unimportant Shinshintô, or a very good polisher might make an mediocre blade look almost like a good one. It takes a lot of knowledge and experience to judge this correctly. An unpolished sword shows none of the criteria that make a sword a good sword except the form, and even that only up to a certain point.

Again I would like to repeat that restoring and polishing has nothing to do with spiritual enlightenment, a Japanese wife or girl-friend, eating sushi at full moon etc. but only with knowledge and experience. One should not attempt to polish a blade without knowing whether it is made by Kotetsu or Kanemitsu. For decades, the NBTHK has supported the training of polishers in order to maintain the traditional art of Togi and to treat the blades with the proper methods. Experimenting with wet stones, sandpaper or the like - even by a professional craftsman - can damage a Nihontô and ruin it forever as the structure and hardness will have changed.

A good polish is achieved by a careful, time-consuming grinding by hand with a series of wet-stones of different consistency and hardness. During this procedure, the bluish and beautiful hada and several tempering-patterns appear. If the tempering was done by a master, nie appears like brilliant tiny mirrors, and nioi looking like clouds during sun-rise. One can easily compare this to ink painting. A master is able to produce a whole color-scale ranging from light gray to dark black on paper. This is much more difficult when the material, as in our case, is made of layers of hard and soft steel.

On the other hand, there are lots of swords that are polished but without deserving this costly treatment; the reason is only that the owner thought that every Japanese blade should be polished. There are people that study for hours a sword that has been polished expensively and which has its origin in a forge of the Japanese Empire during World War II and which shows absolutely no artistic features. The same can be said about many blades from the Muromachi period which are highly esteemed by many Western collectors just because they are Kotô. Many of them are mass-produced, too, and don't reveal anything which could be called worth being collected.

And another big part of Western collections are blades produced by unimportant smiths, showing lots of forging faults, nondescript in form, hada, and hamon, or just boring because "weapons of Japanese origin" were collected instead of "art swords." All the swords of this lowest level are of course not to be classified accordingly, since the marks of schools, times, provinces, or even masters can not even be suspected. Certainly many of these blades show a kind of "quality;" it is even possible that when looking at and examining the blade intensively certain few details can be called beautiful or perfect, but nevertheless, one should be advised against being occupied with blades of low level since bad swords spoil the eyes, figuratively if not literally.

Some collectors seem to have a positive propensity for choosing those types of swords that are best classed as non-Nihontô. Whatever they are, they're not true Nihontô. They are utterly devoid of any artistic feature. Our misguided friend exhibits his non-Nihontô "treasures" with such obvious pride and pleasure that one is placed in a quandary between insipid pretense and brutal honesty. Perhaps the better course is to avoid outright condemnation and to attempt a patient explanation of the basic requirements of a good Nihontô and a gentle comparison of his selections with those preferred by recognized experts. If the explanations and demonstrations fail to register after a few efforts, and irritation and frustration begin to mount, it may be best to desist and to accept the situation. The collector loves his monstrosities faithfully despite confrontation with genuine examples and rational explanations. In such cases further insistence would appear to be a deliberate effort to undermine his pleasure. He's entitled to the protection of the maxim of the ancient hedonists: "If the pleasure is equal, pushpenny is as good as philosophy." Our collector of non-Nihontô has one advantage: his swords usually cost substantially less than sophisticated choices.

To summarize, if Nihontô are worth the money they cost, they should be worth the time and effort required to understand them. Learning Nihontô, like learning any art form, is a gradual accumulation, a slow development of visual and critical acuity, a crystallization of standards, and finally complete rapport with the subject. There is no magic formula and no secret shortcut: the road is tortuous and long....

The benefits of study may be perceived on three levels. At the basic level, the collector learns the distinguishing characteristics of a Nihontô: sugata; hamon and hada and how to recognize them; types and classifications; origin, development, and decline. He begins to distinguish old from new, genuine from copy, crude from fine, commercial from art and other minutiae. He may make some poor choices, but he will learn to rectify his errors.

Often dealers, and some collectors, too, advise neophytes in maxim form: "buy your experience." It's a variant of "learn by your mistakes." They mean by this that the toll for mistakes exacted by the purse makes the most unforgettable lesson of all. The advice is tinged with cynicism. It is true, of course, that experience is a great teacher and we must all learn from it, but there is no wisdom in buying first and discovering the mistake second. As the Chinese sages reasoned, the experience by which one learns need not be one's own. One can learn from the experiences of others and save oneself costly errors. The capsule advice of the numismatists "buy the book before the coin" is much sounder advice. The coin book distinguishes the genuine from the counterfeit and gives dates, identification marks, and values. The coin collector avoids mistakes at the small cost of the book and the time to study its pages. In the same way, the cost of a good library on Nihontô is in most cases much less than that of the purchase of one Nihontô that was priced for fine quality but was actually inferior.

At the intermediate level, to which serious study and application should bring the collector, he will be better informed than the majority of antique and Asian art dealers, except for those few who are specialists in Nihontô. There is no exaggeration in this declaration. For most dealers the Nihontô is only one item among a large stock of numerous categories. Many don't even pretend to be knowledgeable, except the less-reliable ones, whose expertise rises in direct proportion to the ignorance of the customer and the price of the blade. The intermediate collector knows after the first few comments whether the dealer is knowledgeable, and the extent to which his statements may be relied upon.

The advantage to the intermediate collector who knows more about Nihontô than the dealer is considerable. The inexpert dealer will often price his stock unevenly, particularly when he has acquired a collection as a lot. The collector enjoys the possibility of acquiring better Nihontô at lower prices. The intermediate collector is not overawed by the dealer, and at auction sales, he makes his choices and fixes his limits confidently.

At the third level the collector is an expert in his own right. His standards and opinions, developed over a period of time, are his own. He knows that there will always be controversial swords, conflicting opinions, disparate valuations, and questionable standards, but he is also aware that disagreements among experts are limited to a tiny proportion of swords. In the larger areas there is near total accordance among experts. Therefore an opposing opinion, however vehement, will not unsettle him. When he shows a fellow expert his latest prize, which he felt lucky to acquire for $12,000, and the collector says with admiration "What an exquisite blade! If I owned it, you could not buy it from me for less than $5,000!" his feelings are not ruffled.

The expert's standards may or may not accord with popular standards. He may judge a favored Tôshô as overrated. He may regard certain hamon as more indicative of temporary trend than of permanent quality. On the other hand, he may recognize some quality in an unappreciated Tôshô. He may collect certain overlooked Nihontô that according to his standards are undervalued. If his judgments are established in the crucible of time, his bargains of yesteryear may be the record breakers of next year .

The dealer (including the investor) is in search of profit on Nihontô; the collector in search of Nihontô for their own sake. This is basically true. It does not follow, however, that the collector is unaffected by commercial considerations, or that he's unresponsive to the valuations of the marketplace, or that his desires for acquisitions are uncontaminated by extraneous influences. The fact is that it's extremely difficult, if not impossible, to practice a selectivity motivated solely by pure love of beauty. The soul-searching honesty of immaculate aesthetics is beyond most of us. My point may become clear when we attempt to distinguish between the intrinsic and extrinsic qualities of a Nihontô.

What is intrinsic or inherent in a Nihontô is the technique of the craftsman, including his material, combined with his creativeness as an artist. Under strict interpretation of the intrinsic concept of beauty, this quality invests the object once for all when it leaves its creator's hands. After that, everything else that befalls it is extrinsic and subject to the dictates of the marketplace. Under this rigid definition, age and patina of the nakago are not part of intrinsic quality, since they were not wrought by the artist but by time. Even the mei may be considered extrinsic, since it is an identification only and is not part of the design nor of the creation of the essential object. The same is true for the style of yasurime.

Admittedly, so strict an interpretation of intrinsic quality is hard to swallow. A contrary case may be easily made out for the signature as an artistic embellishment, and for age as a result of the period when the artist worked. But even though we accept signature and age as intrinsic qualities of the object itself, we have still to contend with a formidable array of extrinsic influences. Among these are the status of the smith; his popular appeal; whether he or his style is in disfavor or riding a vogue; the region in which he worked; the collectors who owned the piece previously; its use as illustration; the opinions of dealers, critics, and connoisseurs; whether similar pieces have appeared on the market recently and the prices they fetched; the condition of the Nihontô; quality of polish, habaki and shirasaya (or koshirae, which is a field of collecting by itself), and a host of others. One or another of these extraneous considerations may not influence some collectors as much as others, but they do affect all of us to some extent. In any case they certainly influence the valuations that the market places on Nihontô and consequently the prices we must pay for acquiring them.

There are many collectors whose first question is "who signed it?" for an indication of value rather than "how good is it?" for an indication of quality. The concern with signature often seems unduly exaggerated. Some collectors search for a signature before they look for quality. This cardinal concern with signature before quality is certainly a warping of values and a reversal of artistic priorities. Our first consideration should be for the elements of quality. Without quality, the greatest signature is a worthless endorsement. We should be buying the sword, not the name.

The approach of some collectors is even more absurdly tilted toward signature: they are concerned about the very fact of a signature - any signature - with hardly a regard for the identity or status of the Tôshô. Their first question is "is it signed?" not even "who signed it?"; if it is signed - no matter whose the signature may be - it is thought to be somehow better. This is certainly a case of buying a signature instead of a Nihontô.

Not all collectors demand signed Nihontô with an eye to market values. For many a collector, the value of the signature is the assurance of authenticity that it affords. Any misgivings he may have, any lack of confidence in the opinions of others or in his own judgment, these are submerged in the security of a signature. The signature, like the endorsement on a check, tends to bolster his confidence and to support his judgment against whatever doubts there may be. It serves him as a warranty or a certification.

On the other hand, the collector who chooses the unsigned twin at a reduced cost demonstrates a confidence in his knowledge and in his recognition of quality. Being primarily concerned with quality, he readily chooses unsigned Nihontô at a lower price than equivalent Nihontô bearing signatures at a higher price. What is even more advantageous for the confident collector is that he is unlikely to be deceived by false signatures or, regardless of signature, by Nihontô of dubious quality. He relies on quality, not on signature. He relies on contents, not on labels. For him, a signature may confirm or corroborate craftsmanship and style, but he will never accept it as a substitute for quality.

When a distinction between collectors and dealers is based on their appreciation of Nihontô, it is subject to considerable skeptical erosion. There are collectors who know the price of everything and the value of nothing. There are collectors who will not make a purchase unless convinced it is a bargain. There are collectors who practically demand an assurance from the dealer of a prospective increase in value. There are collectors whose proudest boast is for the Nihontô they acquired through their bargaining and cunning, or through the dealer's ignorance. No criticism is directed at collectors who carefully husband their meager allocations in seeking good value for what they spend. The criticism is for those collectors whose primary concern is with price and who search for bargains to an extent that distorts or warps their standards, with the result that quality is relegated to secondary status behind the bargain.

On the other hand, there are dealers who cherish each and every Nihontô that passes through their hands, whether arriving as a purchase or departing as a sale. There are dealers who will deliberately overpay for Nihontô they love, knowing how doubtful it is that they will be able to dispose of them profitably. There are dealers who handle and study their Nihontô with as much tenderness and appreciation as collectors bestow on their favorites.

Sometimes swords have flaws or are damaged. Kizu are of frequent occurrence in Nihontô. It would be surprising were it otherwise. Nihontô were in use for generations over hundreds of years, subject to damage, combat, negligence, and excessive polish. Polishing always involves removal of the original material. Collectors' responses to kizu are as varied as their responses to the Nihontô themselves. Reactions are frequently uncertain, unsettled, or confused, and when they are not, they may be strongly biased. The same collector who accepts one type of kizu with equanimity may shudder with the horrors from another. For some the devotion to a flawed but otherwise fine sword may be as undiminished as a mother's love for her ugly child.

The collector says "I wanted the sword, but it had a kizu." Behind the statement, one senses a quandary, a dilemma, a perplexing question that takes a variety of forms when prompted into frank expression: "Is it wise to buy a damaged Nihontô?" "Can a flawed Nihontô enhance a collection?" "Does the reduced price justify the damaged acquisition?"

An attempt to respond forthwith to the general question with general answers will often disintegrate into compounded confusion and vague generalities lacking content. The problem is that more than one factor is involved. No answer may be final or entirely satisfactory, but certainly the most sensible beginning is to clarify the factors and so narrow the issues. To gain an understanding of some of the factors may help those of us who suffer from "damage vacillation," even though we may not be provided with categorical instructions. It goes without saying that the following doesn't apply to Shinsakutô (modern-era blades).

When is a Nihontô a flawed Nihontô ? In Oriental art, whether bronze, jade, lacquer, Ukiyo-e, or porcelain, condition plays an important part in determining value. Although Nihontô are subject to the same standards, some modification of the definition of perfect condition is applied in their particular case. For example, Inrô or Ukiyo-e in absolutely pristine condition would justify topmost prices, while the identical state in Nihontô, as though fresh from the Tôshô's smithy, would generate suspicions that would undermine a topmost valuation. An antique Nihontô in unused, or rather unpolished, state is, in the eyes of the marketplace, simply unbelievable. Nihontô must be secondhand. They are expected to show some signs of wear. But the wear on a Nihontô must be slight and not so far advanced as to suggest imperfection. In other words, the signs of use must ideally stop at the stage that supports authenticity and go no further. The demarcation between wear that authenticates and enhances, and wear that is detrimental and detracts is fixed - usually but not always - with a fair amount of agreement among collectors. Nevertheless, the point is clear that perfect condition in a Nihontô is not the same unused and unworn state as perfect condition in lacquer, prints, and other categories. In relation to Nihontô, we must accept the Japanese verdict that only Kannon, the goddess of mercy, is perfect.

What is the relationship between Kizu and age? Most collectors will agree that minor jizukare, or kirikomi (battle marks) on the mune, and similar small defects are acceptable in direct relationship to the age of the Nihontô. For example, a hamon narrowed down through repeated polishing in a thirteenth-century Ko-Aoe Tsuguyori blade would be regarded much less seriously than it would be in a seventeenth-century Tsuda Sukehiro, and in the Sukehiro less seriously than in a nineteenth-century Taikei Naotane. The average collector is more lenient in his demand for excellent condition in judging older Nihontô than in judging younger ones. The identical flaw ignored with equanimity in an Bizen Motoshige may cause the ready rejection of a sword by Suishinshi Masahide. This attitude is certainly rational and logical. The older a Nihontô is, the longer it has been exposed to actual use and subjected to polishing, the more wear and injury it may be expected to have sustained during its lifetime. The doctrine of fair wear and tear applies.

What is the relationship between damage and value? We are bound to recognize the unyielding decree of the marketplace that a Nihontô commands a considerable premium for its excellent condition. Perfect state means peak price. Related corollaries are equally true. Small damages will reduce the value of the identical Nihontô, in some cases quite considerably. If we increase the damage to major proportions - hagire (cracks in the edge), fukure (blisters), kakedashi (the hamon extends all the way to the cutting edge), saiha (re-tempering), an obvious repair (like umegane) or a major re-shaping - we may reduce the same fine Nihontô we have under consideration to the price of a mediocrity.

And then, of course, we have "flaws" that are no flaws. A neophyte might reject a sword as being saiha because of mizukage, not knowing it is a common occurrence in Horikawa and Hizen Tadayoshi school swords. He thinks that yakiotoshi, where the hamon begins above the hamachi, means a blade is worthless because he doesn't know it might indicate a sword from Kyûshû.

Many a collector boasts that he will not buy a piece unless it is in perfect condition. Should he discover damage of which he was unaware, he disposes of the Nihontô despite its merits. There is no fault to find with the collector who demands perfect condition and is willing to pay peak prices. The perfectionist is quite understandable. The art advisers approve his attitude as an investment policy. They assert that it is the art objects in the best condition that show the greatest increase in price. However, investment appreciation is not every collector's concern, and not every collector can afford premium prices.

At the opposite extreme from the collection of perfect specimens, or Perfect Collection, as we may call it, is the collection of damaged Nihontô - the Damaged Collection, so to speak. A collection of damaged pieces would certainly be acquired on the cheap, though it might well contain, however dismaying the damage, a sprinkling of great Tôshô and great forging.

As a brief diversion, it is amusing to imagine a collector offering examples from his collection to illustrate not smiths and schools but types of kizu! As bizarre as this might sound, collecting flawed Nihontô is not unimaginable, nor is it necessarily an absurdity. The collector who acquires damaged pieces is not ipso facto a radical or a dolt.

If I may deduce a principle from the above, it is that a Nihontô may have flaws yet retain its uniqueness and original quality, or it may suffer minor damage that severely compromises its merits. Every collector must answer to his own satisfaction to what extent the ugliness of the flaw mars the beauty or rarity of the piece. Despite all that may be said in order to clarify the factors and to fix guidelines, the ultimate decision about the acquisition of damaged Nihontô is no less subjective than it would be were the Nihontô undamaged.

Somewhat tangentially, I would like to take note of the collector who prides himself on his eagle eye, his inordinate ability to spot the most minor defect or damage. A Nihontô one has owned for ages and has examined numerous times may suddenly reveal a defect. When we ourselves discover a tiny scratch after a long period of ownership, we suffer the remorse of a confessed vandal in fear that we ourselves, by our own carelessness, may have caused the damage. We ask ourselves whether we could have failed to notice the damage when we acquired the piece and at all the other times when we examined it. More often, however, it is our colleague of the piercing vision who spots the injury. Our emotions on being advised of damage to a heretofore perfect sword are of course mixed. We like to know all we possibly can about our own Nihontô, but the damage dismays us. Besides, we would like to discover the faults ourselves and not be told about them by a visitor who was only supposed to look.

Occasionally someone will point out to us alleged damage that, to our great relief, we find is fancied or imagined or that results from the spotter's ignorance. We sometimes experience a tinge of annoyance with this expert finder of damage, this defect fetishist. Our annoyance with the damage-spotter burgeons if we know that detecting damage is his only claim to expertness. He boasts about his sharp vision. "No damage gets by me," he says. In his inordinate, abnormal concern with damage, he is like the specialist in pathology examining diseased tissue under a microscope, or the monkey searching for lice. We do not mind so much if our owl-eyed friend is genuinely interested in Nihontô. The damage-spotter we resent is the one whose expertness is limited to imperfections and who has very little knowledge of Nihontô in general. He needs no reading, no study, and no application for his gratuitous field of endeavor. He needs only visual acuity and a morbid fascination with defects. He can exercise his exaggerated concern free of all sensitivity to craftsmanship and artistry.

The knowledgeable collector who looks for quality will not be influenced by great name or by snob appeal. He may prefer the dealers' auctions, feeling that there he will more likely find bargains. When he attends the name auctions, he will bid for the unnoticed or overlooked sleeper, but he will retreat from the contagious contests for popular masterpieces that grow too heated. This course is not in conflict with the excellent maxim "buy the best quality you can afford." The advice, when expertly trimmed, is to buy the best quality, but not necessarily to look for it where it is popularly assumed to be, nor to acquire it at absurd cost.

Quality Nihontô at bargain prices may sometimes be found in the shops of dealers. Not every dealer is a Nihontô expert, but even the expert dealer is inevitably influenced in fixing his prices not only by his costs but also by his prejudices and preferences, his personal likes and dislikes, and his arbitrary opinion about what a particular sword is worth. It is often said that art objects, including Nihontô, are blind or half-blind. The "blindness" is in the subjectiveness of value, the impossibility of determining a precise value or of scaling uniform increments for higher or lower quality. Nihontô cannot be precisely priced like a box of cereals.

To warrant a top valuation, a Nihontô must accord with the style that is regarded as characteristic of the Tôshô. It's the smith's characteristic work - the unmistakable work - that arouses the stiffest competition. Is this standard justified ? The matter is certainly controversial. The affirmatives say that Echizen Yasutsugu's swords with Aoi-Mon were forged in his golden period and are worth more than his swords from Shimozaka days. However, a few observations may be relevant. As a creator, the artist sees the world with fresh eyes. He may lose his "innocence" when under social pressure, after becoming famous and expected to make swords exactly like those that brought on this fame in the first place. On the other hand, his long experience gives him a control over the blade he didn't have when he started forging swords.

Fads, trends, and vogues do not exert the sudden life-and-death effect on Nihontô that they do on the products of modern art. However, interest in a certain Tôshô or a particular school tends to flare and surge now and then in direct reaction to a record price or to an important sword. In those cases where overlooked and undervalued Nihontô are spotlighted to reveal their true worth, the increase in price is most proper . The new price accords with quality. In other cases, however, the response is not so much to quality as to the tendency to join popular movements, to go with the crowd. Fads, vogues, and trends in art come and go, but quality and good taste endure. The collector who assiduously searches for quality rather than for the "in" Tôshô or the trendy Bizen model will in the long run pay less and have more. Vogues and trends are temporary, quality is permanent.

Now to deal with our main theme, lawful avoidance: how to continue collecting while avoiding error, extravagance, and excessive cost.

As he shows his acquisitions, a collector sometimes makes one or another of the following statements somewhat apologetically:

"This sword isn't very good, but I thought I ought to have an example of this school."
"I bought this one because I needed the naginata-naoshi type."
"I've given up trying to find this smith in decent quality; I was lucky to get this one at all."
"I bought this for Hamon, since I didn't have saka-chôji."
"I suppose every collection should have one of these."

Why should any collector feel compelled to buy a Nihontô when his heart isn't in it? To the extent that he makes purchases of things he doesn't really want, he is not collecting but accumulating. He is not expressing himself, his own interests, his own tastes and preferences, but those of someone else. He is collecting by rote and by popular demand, by what he feels is expected of him and of his collection. If this kind of collecting is carried to the extreme, to the ultimate reductio ad absurdum, he makes a collection for everyone else but not for himself. What is the purpose of a collection if not to delight the owner and to make a personal statement of his interest in the art?

There is another aspect of the "ought-to-have" collection in lieu of the "what-I-like" collection. If one acquires Nihontô because they fill voids, blanks, and gaps in his collection and continues on this course, he will as a logical conclusion accumulate examples of all types, hamon, periods, areas, and artists. His collection will eventually be fully representative and will encompass the universe of Nihontô. It will be a Comprehensive Collection. It is the type of collection many collectors easily amassed during the plentiful and cheap era of Nihontô collecting before the 1960's.

The Comprehensive Collection is best suited to function as a study and teaching collection. It includes examples to illustrate almost any point about Nihontô that the student might wish to learn or the teacher might wish to illustrate. Since it covers all phases and facets of the subject, it is necessarily of considerable size. It is the proper collection for the museum or institution. Comprehensive Collections, like other collections, may be found on various levels of quality from ordinary to superb. Quality would have little effect on its usefulness as a tool for learning and teaching. What all Comprehensive Collections have in common, regardless of quality, is their horizontal formation. They cover the entire field - a characteristic that gives the collection, in geometric terms, its level plane. It need not deal with any particular phase in depth. It therefore stands in contrast with the vertical line of the in-depth collection, which is specialized or limited to particular aspects. The Specialized Collection will be discussed later.

There is certainly no fault to find with the collector whose interest and predilection lead him to the Comprehensive Collection. It is a commendable goal if it is his pleasure to represent fully the various aspects, phases, and facets of Nihontô. He can explain and demonstrate any direction the Nihontô compass shows. The point is that our comprehensive collector is then expressing his interest, he is "doing his own thing". The objection is to the practice of the collector whose haphazard acquisitions "by popular request" are leading him willy-nilly in the direction of the Comprehensive Collection, which he might be the first to disclaim as his goal. To the extent that he buys what he neither needs nor likes, he is committing the utmost extravagance. He pays for third-party approval and wastes his money on Nihontô that do not satisfy him most of all.

Save the money that you would spend on Nihontô you ought to have, and spend it instead on the Nihontô you want to have. Take the piece that satisfies your heart's desire, and reject the one that leaves you unaffected, even though it be the object of great acclaim. Your acquisitions should elate your spirits; they need not stir the envy nor win the approval of your colleagues. As a sincere collector, you do what you should when you listen to new suggestions and contrary opinions patiently, and consider them thoughtfully, but extract only so much as commingles naturally with your own taste and bent.

Some collectors want to have nothing but "the best". Whatever acquisitions they make must be "best of class." They aim at a collection that will be of zenith quality in each and every sword. It is hardly informative to point out that the competitions for the best Rai Kunitoshi, the best Hankei, and the best Gassan Sadaichi are the hottest and costliest. Winning bids are always at record levels, and the collector must be prepared for an exceptionally high unit cost. The collector in pursuit of such rare examples joins the hounds in pursuit of the fox - a small pack, true, but way too large for the rarity of the quarry. The number of Kunitoshi blades is limited, but the number of collectors is virtually unlimited.

The first question to ask about such an ambition is whether the goal of the Best Collection is attainable. The answer is a clear "no," for the reason that what is best in art is neither definable nor measurable in absolute terms. On the contrary, it is variable and subjective. While our collector of the best will not acquire the Best Collection, he may acquire a superb collection, one that is superb in the unanimous opinion of all connoisseurs. However, some collections of superb quality have attained the rank despite the absence of a single Rai. In this respect, collecting Nihontô does not differ from collecting Ukiyo-e, postage stamps, rare books, or Chinese porcelain. There are great collections of Ukiyo-e that do not contain an unpublished Sharaku, of postage stamps that do not contain a British 1840 Penny Black, of rare books that do not contain a Gutenberg Bible or a Shakespeare folio edition.

Possibly the reason for the sense of deprivation of the collector who finds the "best smiths" beyond his grasp, is his proximity in time to the easy availability of these rarities prior to the 1960's. Although fine examples have been acquired by many living collectors, the gap in a collection is no justification for frustration or despair. Even without a representative work of Kotetsu, superb collections not only are in existence but also are in a nascent stage of formation.

There is a characteristic of the Best Collection that should be pointed out here as an aid to an understanding of the Specialized or Vertical Collection, which we will discuss later. The Best Collection is on the rarified level of peaks, pinnacles, and summits. It does not descend to plateaus, foothills, and valleys that stretch out between. It is an assemblage of superb examples without any particular relationship except for the fact that all are Nihontô. The collection will demonstrate that Nihontô are marvels in steel. It is, however, lacking in standards, comparisons, and relationships; it is lacking in depth. In geometrical terms, the Best Collection resembles the previously discussed Comprehensive Collection. Both rest on horizontal planes, although the Best Collection attains the more rarified atmosphere of a high plateau.

Charles Mitchell suggested a type of collection that he applied most successfully to Ukiyo-e, a field in which he is a recognized expert. He calls it the Quintessential Collection. He describes the planning and composition of the Quintessential collection in substantially the following words: "I had a predetermined limit of 20 prints with a predetermined breakdown according to periods, artists, subjects, and formats. Approximately 25% of the prints were primitives, 5% in the middle period, and 25% in the late period. Each of the six most important artists had to be represented by at least six prints. All other major artists had to have a proportionate representation, and interesting minor artists had to have some representation. The breakdown by subject was 3% actors, 25% beautiful women, 15% erotica, 10% genre scenes, 10% historical subjects, and 5% each birds and flowers and comic prints. All of the various formats and techniques had to be covered by the collection in about the same proportion as they occurred in the art itself. It was fun plotting all this out, and it certainly posed many problems in the course of collecting to fill the blanks, but in the end I had something that could be called a Quintessential Collection, and a viewing of the 20 prints would give someone an accurate idea of what Ukiyo-e was as a whole."

The Quintessential Collection is easily applicable to Nihontô. Instead of Ukiyo-e formats we would have Tachi, Katana, Wakizashi, Tantô. Instead of primitive prints we would have Kamakura period blades. Instead of major and minor artists we would have a range of Tôshô in our examples. Types of hamon would be extensively represented. The Quintessential Collection is quite well-adapted to Nihontô. It offers the prudent collector important advantages.

Before the collector can embark upon his quest for the Nihontô that will compose his Quintessential Collection, he must determine how many pieces he will acquire, which aspects of the art are major and should be emphasized, which aspects are minor and need only thin representation, and finally the percentages he will devote to each facet of the art so that it may be viewed in its entirety. His acquisitions are deliberately plotted and planned in advance with pencil and paper. As the collection takes form and purchases are made, the spaces on the plot are checked off. Thus the Quintessential Collector must be prepared with prior knowledge of the art and with a thoughtfully-deliberated plan. He must know every facet in order to make intelligent decisions about his ideal representations and percentages. The economical benefit of familiarization and study of the subject prior to making acquisitions is obvious. Our largest percentage of errors will usually occur in our initial burst of enthusiasm when we are introduced to the art, and each and every Nihontô seems wonderful, fresh, and free of flaws.

The Quintessential Collection, by its very nature as a deliberately planned operation, tends to eliminate certain collecting errors. Many collections grow by haphazard accumulation. The collector buys a sword because it is attractive or of good quality or a bargain or because he feels rich that day. Although he consistently buys good quality, he may find himself loaded with more of a particular type than he needs. His acquisitions are fine, except that they are sometimes duplications or repetitions. Like the magic bean stalk, his collection just grows and grows with no apparent limit in size or balance in representation. A fixed numerical limit and a predetermined plan accord purpose and significance to each acquisition. The plot and plan of the Quintessential Collection requires that each piece shall fill a space and represent the specific aspect of the art for which it was intended. It must "fit" as an integral unit of the entire representation of the art. One Nihontô may serve to represent several different aspects of Nihontô. For example, it may represent Tachi as a type; chû-suguha as a hamon; ko-nie and nioi as habuchi; ashi, yô, kinsuji, and inazuma as hataraki; a bo-hi as horimono; rai-hada as jigane; and Ryôkai as a Tôshô. Obviously there should be no waste, no misspent money, and no duplication in a Quintessential Collection.

Even though limited to 10 or 15 swords, the Quintessential Collection may be sufficient numerically to represent accurately and adequately the art of Nihontô as a whole. As Nihontô increase in scarcity and expensiveness, the smaller collection becomes more rational and practical. Collectors may find the Quintessential Collection, with its built-in safeguard against error and waste, a prudent and economical method of collecting. It is never too late for the collector to turn the direction of his acquisitions into a Quintessential Collection, though it will require the formulation of a game plan and the elimination of Nihontô that do not belong.

The Quintessential Collector must have the tenacity to adhere to and persevere with his plan, and to refrain from acquisitions, however alluring, that do not fill the voids in his plan. When his collection is completed, he will have the satisfaction of showing a relatively small collection that is nevertheless representative of the art as a whole.

The Specialized Collection stands in geometric contrast with the horizontal forms of the Comprehensive Collection and the Best Collection. Its form is vertical. It is the in-depth collection. It is the favored collecting approach of connoisseurs whose axiom is "collect in depth." What does the term mean, and why is it wise to "collect in depth?"

We begin with analogies from the fields of postage stamps and Chinese porcelain. These hobbies have stood the tests of greater numbers and longer time than Nihontô. The philatelist soon learns that amassing a few stamps from this country and a few from that, a few of this period and a few of that, a few of one type and a few of another, and a few misprints makes a poor collection. It is an assemblage of unrelated specimens, except for their common classification as postage stamps. It is the juvenile and amateur approach to philately. The knowledgeable collector specializes, whether in French colonial stamps or nineteenth-century Canadian stamps or floral stamps or triangular stamps, but he always collects according to his special interests. Likewise the thoughtful collector of Chinese porcelain does not assemble one example each of various periods, areas, types, and glazes, but instead concentrates on Ming blue-and-white or Qing export ware. This is the in-depth collection or Specialized Collection. It tells the story of a particular aspect, its origin and development, its flowering and decadence, its evolvement, modifications, and offshoots. It shows standards, relationships, and comparisons. It informs, demonstrates, and teaches. We learn from it and expand our knowledge. Between the high walls of its narrow edifice, it reveals the whole story of a specific aspect of the art. It is an eminently effective collection. Each member of the Specialized Collection is related in some way to every other member. Each unit gains something from the other units of the collection. Thus the collection as a whole is worth more than the total of its individual pieces. It exemplifies the maxim "a good collection is greater than the sum of its parts." It forms a true collection and it is more valuable as a collection than the totality of the units of which it is comprised.

How may the principles of the Specialized Collection be applied to Nihontô? Let us suppose that a collector notes a few pieces signed Ômi no Kami Sukemasa (the name of any other little-known Tôshô will serve as well). He is attracted by Sukemasa's strong jigane, his thick, even nie, his healthy functionalism. He detects a resemblance to Inoue Shinkai, and learns that Sukemasa was one of the "hammer men" of the hobby smith Tokugawa Nariaki. He gradually comes to the conclusion that Sukemasa is a fine Shinshintô-period Tôshô relatively unnoticed and unreported. He feels that he is not appreciated at his true worth. He decides to collect Sukemasas. By assiduously watching the shops and auctions over a period of time, he gathers together an impressive number of his swords.

What has our hypothetical Sukemasa collector accomplished?

1. He has amassed a unique collection. In numbers and in depth, there is no comparable collection of Naoe Sukemasa.
2. His collection affords the basis for a survey and history of Sukemasa as a Tôshô: his technique, treatment and style, his first efforts, development, best period, and late work; his ability and standing; his peculiarities and singularities; his successes and failures, how he compares with other Tôshô of similar style in the same period, and his relationships to other Tôshô as teacher or apprentice or co-worker.
3. He has made a contribution to our knowledge of Nihontô. He has filled in some of the gaps and voids in our information about Sukemasa. His collection is instructive. It arouses our appreciation and respect.
4. He has revealed a good deal about himself, his personal relationship to Nihontô, his individual interests, preferences, and taste.
5. He has accomplished this at low cost. He collected Sukemasa when his merit was unrecognized and at a time when competition for his work was low and limited. Moreover, he acquired not only Sukemasa's "golden period" but also his early work and late work; not only his signal successes but also his telling "misses." As a "complete" collection, his Sukemasa are more valuable than the sum of the individual pieces. If his judgment about Sukemasa's quality is correct, the marketplace may eventually reflect an increase in its valuation of Sukemasa, much to the credit of his "discoverer."

The advantages of the Specialized Collection are considerable. It is a collection that is inspired by the collector's individual interests, tastes, and predilections. It accords with his personality and character, even with his whims and idiosyncrasies. It is his collection, based on his choice of those aspects of the subject that he finds most appealing and interesting. An important advantage of the Specialized Collection is the cost, which will be much lower than the cost for an assemblage of choice but unrelated examples. Since the purpose of the in-depth collection is to tell a complete story, it will include mediocre examples at low cost, as well as the Tôshô's superb flowerings at higher cost. Many specimens and examples, although important for standards, comparisons, and relationships, will cost considerably less than unrelated rarities. The Specialized Collector is apt to find himself a "loner" in the pursuit of his goal. He faces minimal competitive interest from the mass of collectors. Despite the lower cost, the interest and effectiveness of the Specialized Collection may be greater than that of some Best Collections.

Perhaps the most exciting prospect open to the Specialized Collector is that of uncovering a Tôshô, a technique, or a type that for one reason or another is unnoticed, neglected, ignored, or unappreciated. He has the opportunity to pursue his natural partialities and preferences, and at the same time to focus attention on some hitherto-overlooked group of Nihontô. A considerable volume of literature has been published that has led directly to reappraisals of individual Tôshô or schools.

This doesn't mean that the last word has been spoken about better known swords, like Hizen-tô. R. J. Robertshaw showed us in 1999 with his book "The School of Tadayoshi" how original research and a fresh approach can make us seeing this lineage of smiths in a new light. The possibilities are quite numerous. They are waiting for their discoverers.

I stress time and again that only studying the subject in depth will enable us to form a collection that is both within our financial reach and worth being actually called a collection due to its quality. But I also have to admit that only reading books and looking at swords without guidance will never lead us to true connoisseurship. We need to learn from our peers as well, be it as a member of a sword club or under the tutelage of a seasoned collector.

I was lucky to both having lived extended periods of time in Japan, and meeting some fine gentlemen who freely shared their knowledge and "taught me swords" without ever asking anything in return. But being the student of a true expert doesn't automatically make oneself an expert, too. The taller the shoulders of the giants on which we try to stand, the deeper and harder the fall can be. I don't say this to appear appropriately Zen-ish, but in the painful knowledge that I probably never will reach the level of expertise of some of those who taught me.

I also met some very strange persons in the "sword world," in Japan as well as in Europe and the United States. It therefore makes learning the basics even more important in order to stay clear of those self-proclaimed "experts" who have an answer to everything and true knowledge of nothing. This is especially true for the Internet. You never know if the person who answers your question is a Nihontô scholar of note, or just some pretender who happens to own a couple of books on Nihontô he uses to the best of his advantage in disguising himself as a connoisseur. Those with experience look behind the facade easily enough, just as a scientist who devoted his entire life to research knows after the first few words whether his colleague is exactly that, or just someone who answered the advertisement, "Diplomas from prestigious non-accredited universities, no required tests, classes, books, or interviews. Bachelors, masters, MBA, and doctorate (PhD) diplomas available in the field of your choice."

Knowledge can't be bought like the above-mentioned diplomas. Only the newcomer might be impressed by the aplomb, the liberal use of Pidgin-Japanese, and the real or pretended acquaintance with well-known collectors and/or dealers our hypothetical impostor uses to corroborate his "credibility."

So who are the real experts? Some collectors want to make you believe that only a person of Japanese descent, "by grace of birth," so to speak, can gain deep insight into this complex field of studies - a myth many Japanese enjoy supporting. Granted, experts of a caliber of Dr. Homma Junji or Satô Kanzan are not found outside Japan. And although it's certainly true that a Japanese has the advantage of easier access to information - and art swords - it finally all comes down to the effort one puts into his studies, not place of birth or DNA. Quite a few non-Japanese have proven that in the past and will continue doing so in the future.

The critical reader may - not without reason - accuse me of having made myself the judge of what is right and wrong in our approach of appreciating and collecting Bijutsu Tôken. In writing law, a judge is limited to what is necessary to decide the case. Anything beyond what is necessary is termed "obiter dicta" and is not binding. The saying is that judges decide cases; they do not write law. Nevertheless, a judge is occasionally carried away with the force of his reasoning, and applies it to collateral matters not at issue. I hope I will be forgiven for having inserted here quite a few obiter dicta of my own. Like the judge's, it may be good law or it may be bad law, but since it is not binding, it will injure no one.

"The typical collector is a male, usually unbalanced in some direction, who, if he were normal, would not need to collect odd bits and pieces. I believe that any collector suffers from some kind of mental or psychological aberration, and that his collecting is a therapy which may run into a great deal of expense but which protects his sanity and allows him to operate in other fields fairly normally."
James A. Michener


Stasis in Complex Artifacts: Nihonto

Most of this paper is concerned with mechanisms for the generation of new knowledge, particularly the type of knowledge of the world around us which helps us to create new resources.

During the long history of Homo sapiens - a timeframe subject to much debate, but certainly in excess of 40,000 years - a great deal of resource (product) innovation has taken place, embodying much new tacit and reliable knowledge, but the rate of innovation has been far from constant. Periods of change have been punctuated by long periods of stasis, so that for most of history, the world would seem to be pretty much the same when a person died to the time when they were born.

This phenomenon, of high rates of innovation followed by stasis extending over many generations is not confined to the less-developed, hunter-gatherer or early farming societies, but is also exhibited in advanced and sophisticated civilizations which had learned to beat the Malthusian trap, to produce resources to provision steadily-rising populations.

Japan provides us with one such example of a sophisticated society, many of whose artifacts ­resources - had remained largely unchanged, apart from decorative variation for many centuries before substantial contact with the Western world. We shall examine here one of the most famous of Japanese artifacts embodying extraordinarily high levels of craft skill, but which remained virtually unchanged in form, function, and manufacturing process for over 700 years: the Japanese sword. We will show how the complexity of the manufacturing process, and its sequential nature in which each step is crucially dependent on the meticulous and precise completion of the previous stages tends to produce 'lock-in.' Variation and innovation become increasingly hazardous and are confined to minor decorative features which, while largely irrelevant from a functional point of view, assume great social importance.

Steel swords have been found in tombs in Japan datable to the fourth or fifth century A.D., but it is probably that these were made in China and brought to Japan through Korea. Swords were being made in Japan by the Heian period (794-1185 A.D.) but the special characteristics which made the Japanese sword such a formidable artifact were not developed until the Kamakura period (1185-1333).

The functionality and processes of manufacture remained virtually unchanged from this time until 1876 - eight years after the Meiji restoration - when the warrior class, the Samurai, were formally abolished. Within a century, the sword had become obsolete, except for the small numbers produced purely as art objects and mass-produced swords of much lower quality for military officers.

The Japanese sword is a cutting tool originally developed for use from horseback with a slashing action, not the thrusting action of many Western swords. The blade is typically 22 to 30 inches long, and has a handle suitable for two-handed operation. It is slightly curved, with a sharpened edge on the convex side of the curve. The concave side is not intended for cutting, and is blunt.
The functional requirements are straightforward - that the cutting edge be capable of extreme sharpness, neither dulling nor chipping when violently striking enemy armor of cotton, leather, or iron, and often flesh and bone. The second important functional requirement is that the blade remain straight and in one piece in use, even when struck hard by, say, another sword in battle. There can be few occasions more embarrassing - and deadly - than to be left holding a handle alone.

There are other slightly-subsidiary requirements: the shape of the blade should assist a slicing action; the weight distribution should be such that the sword has plenty of momentum on the one hand, but is balanced and not awkward to use on the other; and the surface is of a composition and finish that does not encourage rusting, as long as care is taken.

The requirements of hardness and toughness don't normally go together, particularly in steels. To remain sharp, an edge must remain very hard - certainly much harder or resistant to deformation than the thing it is intended to cut. But hardness usually goes with brittleness, the opposite of toughness. The strength - literally - of the Japanese sword blade is that it combines these two qualities, hardness and toughness, in a most successful manner, using steel in its hardest form for the cutting edge, while using steel in its toughest form to back-up the cutting edge and to form the body of the blade.

The creation of this combination of properties requires the most extraordinary skill on the part of the swordsmith, and it is the recognition of this level of skill that ensures the blades a special place in our museums, and the accolade of 'Loving National Treasure' to the best of their makers in Japan.

Of all the metals, it is iron which has had the most profound effect on the course of history. It is abundant, comprising around 4.5% of the earth's crust - 200 times more abundant than copper and 3000 times more abundant than tin, the two constituents of the alloy which gave us the bronze age.
Pure iron - rarely encountered even today - is quite soft and bendable, and has very few practical applications. The properties of the metal change markedly, however, with the incorporation of quite small amounts of carbon. As small an amount as 0.4% carbon permits the material, which we call steel, to be capable of heat treatment to a greater hardness than the best bronze tools with which the Egyptians carved their huge granite statues.

Japanese swords typically have an edge of iron containing 0.6-0.7% carbon which permits, with suitable heat treatment, a hardness sufficient to scratch glass. Higher amounts of carbon are used in a variety of applications in Western industry - the massive tools used for molding plastic or forming car bodies may have over 1% carbon, and the 'cast iron' which used to be the standard material for stoves and grates, manhole covers, railings, etc., had up to 3%. The percentage of carbon for any given application, and its distribution in the iron and the heat treatment the iron/carbon material (steel) has undergone, are crucial to the satisfactory functioning of the
artifact - in our case, the blade.

The Japanese knew nothing of carbon - neither did anyone else in the heyday of the sword - it was not identified as a separate material, an element, until the end of the 18th century. Nor did they know that they were adding this all-important material accidentally to the iron during the process of extraction of the iron from its ore, iron oxide. The extraction of iron entails mixing pulverized ore with charcoal, igniting and blowing air through the mixture in a furnace large enough to ensure attainment of a very high temperature.

The charcoal burns to carbon dioxide and carbon monoxide; the carbon monoxide reacts with the ore at high temperature, removing the oxygen and leaving iron. The temperature was not high enough to actually melt the iron which stays in the fired mass as friable lumps, to be picked out when the furnace had cooled down. These lumps were then reheated and hammered repeatedly to produce a usable and workable material.

During this process, hot iron extracted from the ore comes into contact with hot charcoal - which is essentially carbon - and absorbs it quite readily. However, the amount of carbon absorbed cannot be controlled, and the process is quite haphazard, producing a carbon content range from less than 0.5% to over 1.5%. The swordsmith is highly skilled at sorting the lumps of raw steel, called tamahagane, from the furnace and judging from appearance and texture their suitability for the hard outer shell of the blade, including the cutting edge, and the softer, more ductile core. To do this, he takes lumps of the raw, friable steel from the furnace, reheats it to a bright-red heat and hammers it into sheets about a 0.25-inch thick. While still hot, the sheets are dropped into water; the rapid cooling leaves them in a very hard and brittle state. They are then broken up with a hammer to form flat but rough-shaped pieces each with an area somewhat less than one square inch.

The finished sword needs steel with 0.6% to 0.7% carbon at the cutting edge, but many reheats are required during the forging process, and each time the steel is reheated in the forge a little carbon is burned away, along with impurities. For this reason, the smith starts the whole process with steel containing 1% to 1.5% carbon, and the carbon content gradually drops during the long and many-times-repeated heat and hammer, heat and hammer. It is worth remembering here that the swordsmith has no real knowledge of what he is doing in metallurgical or chemical terms. Only in this century has he been able to think in terms of percentages of carbon; ­indeed, he was unable to think in terms of carbon, or some unidentified substance being present in the iron and dramatically altering its properties, at all.

In Europe, for instance, in the middle of the 18th century, it was widely thought that steel was a particularly pure form of iron, purified by the heating process. By observing tiny clues in the process and thoughtfully relating them to the properties of the finished article, Japanese swordsmiths produced a remarkable consistency in composition and hardness.

The little chunks of hammered and broken raw steel are carefully examined at their fractured edges, and those which are heavy and dense, with a fine crystalline structure and a bright silvery color - which we now know to have between 1% and 1.5% carbon and a few other minor impurities - were then known to be the right material to start the blade-making process. They were carefully sorted out and stacked in a pile 6-7 layers high and 3-5 inch on a side, heated to a bright yellow color (around 1300C) in a charcoal forge, and hammered to a solid mass.

Iron and steel in this carbon content range (but not above 1.5% carbon) may be welded one piece to another by heating to this high temperature and then hammering them together. The hammering is done in a manner that elongates the block, and when a long flat bar has been obtained, it is folded over on itself and the process repeated, rather as one might make pastry, but in this case with hot metal and a heavy hammer.

Yoshindo Yoshihara, a modern swordsmith who describes this process in great detail, folds and hammers out the steel about 13 times (producing around 16,000 layers per inch of thickness). The repeated heat-hammer-fold, heat-hammer-fold homogenizes the material, distributing the carbon evenly, while a little is constantly being lost. Eventually the material has reached a stage at which the smith considers it ready to fashion the rough shape of the blade. However, this would produce a blade of uniform material, and a blade more able to resist shock can be produced by inserting a core of a softer more-ductile steel right down the length of the billet

Just as the original broken chunks of raw steel were selected to produce a steel that would give a hard edge, so other chunks are selected - those with a darker, muddy-looking fracture - which have a much lower carbon content. These are destined, after their own heat-hammer-fold processing, to form the softer, ductile core. This is done by chiseling a deep V-cut along the length of the first steel billet, performed when the steel is white hot. The softer core material is forged to a shape suitable for insertion into the V, and the two are at very high temperature inseparably hammered together, the core inside and towards the back of the blade. The composite blade is now reforged into its final shape.

Next comes the final heat treatment to bring out the strength of the blade and the hardness of the edge. The degree of skill required for this stage may be judged by the length of time devoted to it during a swordsmith's apprenticeship. All the preceding stages would be taught in two to three years. A further two to three years would be devoted to the final crucial hardening and the production of a blade ready for, but excluding, the fine grinding of the surfaces and the edge sharpening.

While the presence of a small percentage of carbon is needed to produce a very hard steel, a careful and closely-controlled heat treatment is required. This is done by heating the steel to a temperature above about 720C (which the smith would have to judge entirely by eye), a bright cherry red, and then cooling rapidly in a tank of water, an operation called quenching. If the steel is allowed to cool slowly - just held to cool in the air for instance - it will not develop the degree of hardness required for a cutting edge. Quench hardening has been recognized for many centuries: Roman swords exist that show evidence of quenching - but the underlying theory of just what is happening in the crystals of steel, that gives hardness with fast cooling and softness with slow cooling (dislocation theory) has only been understood within the last few decades.

The swordsmith wishes to make a blade which is very hard at the cutting edge, while much softer and tougher immediately behind the edge. This is quite apart from the use of soft steel in the core.
To achieve this, he ensures that the edge cools very rapidly but that the rest of the blade cools much more slowly, ending up much tougher. The method is to coat the blade with a layer of clay, very thinly applied at the edge, and much thicker towards the back. When the blade is thrust, red-hot, into water, the edge cools rapidly through its thin clay coating, the back more slowly.

After final grinding and finishing of the blade, the difference in color and texture between the hard edge and the softer back steel can be clearly seen, and many centuries ago this dividing line was understood by the client warrior to be good evidence that the blade had been made out of the correct materials and properly hardened.

Over the centuries, swordsmiths learned to slightly contour the clay coating, producing patterned effects in the steel behind the edge. Different schools of swordsmiths, and different individual smiths, would develop different patterns that have become important, though almost entirely decorative, features in the assessment of the sword. If you take a piece of steel, two or three feet long, heat it to a bright-red heat, and suddenly thrust it into a tank of cold water you can expect trouble. The whole blade can bend or twist, or cracks can form through the body or at the surface.
After all the work that has gone into the blade up to this stage, the average swordsmith would expect to lose over half of his work and an extremely skilled smith may lose a quarter.

Some correction of faults may be possible, including a little judicious straightening or correction of curvature, but the loss figures just given would apply after any such corrections had been attempted. For blades which survive this ordeal by fire and water, the long process of grinding and sharpening commences. Again, extreme skill is called for, using carefully selected natural stones in many degrees of fineness and hardness so that the final product emerges with sharply-defined contours, a smooth sweeping curve, and a very finely textured, though not polished, finish.

The swordsmith's skills evolved over centuries. Having evolved, they remained relatively static for many more centuries. Enormous experience and skill has been accumulated in the choice of materials for each stage of the process - the choice and concentration of the original ore, the cleaning away of impurities, the firing with carefully prepared charcoal to extract the first raw steel, the hammering and quenching - all leading to skilled assessment of 'suitability' which we now call carbon content, the refinement by many times repeated forging and folding; the meticulous coating with clay to insulate one part of the blade more than an immediately adjacent part during the hardening and quenching process.

The reader may care to think through this last stage in detail. Imagine that you take a large carving knife from your kitchen and that you must cover it with a coating of clay, as thick as a playing card close to the edge, stepping up to the thickness of half a dozen cards towards the back of the blade. You must scour the countryside for your own suitable clay (you will have to try many), and you must learn to refine it and prepare a thin slurry.

You must then heat the blade with its thick envelope of clay to a bright-red heat over a bed of burning charcoal so that it has the same high temperature right along the length and on each side, all without disturbing the clay. And, when you have made the temperature both even and correct, you must immediately thrust it into a tank of cold water. If any of the clay comes away from the steel, a whole month's work is instantly lost - and there are children with hungry mouths in your house to feed
Having found a clay that works in spite of this violent treatment, you treasure it. You lay hands on enough to last you throughout your career. You take great care to make sure there are no oversize particles to give little drag lines as you apply it with a spatula to the blade, with no minute specks of organic matter which could burn with the heat and lift the layer. You will develop extreme caution in the surface finish of the steel to which you apply the slurry - not a hint of grease, not too smooth, a nice even oxide coating, but not a scale which could become detached.

And when you have it right, you adopt one golden rule; don't change anything. Use exactly the same materials - even though you don't know in any rigorous way what they are - prepare them and process them in exactly the same way each time. Suppress variation, suppress innovation, and train your apprentices to stick rigidly to the rules. This type of complex, sequential process leads almost inevitably to lock-in, to stasis. Even when using the best and most careful techniques, a high failure rate occurs, and the only way to achieve a success rate that can be lived with is to repeat each stage as exactly as possible. Innovation is extremely risky and is likely to lead to expensive failure.

The Japanese sword has been used as an example of this phenomenon, but there are many others. Particularly susceptible to lock-in are processes in which slight changes in one ill-understood variable produce large changes in the product. This occurs frequently, for example, in early glass manufacturing, in which the absence of a small percentage of calcium oxide (accidentally incorporated in Roman glass by shells in the silica sand deposits used as the raw material) leads to an unstable, water-sensitive glass.