Hoshi

There appears to be some misunderstanding: I am writing about smiths as statistical distributions. Of course, each sword must be judged on its own merit. But If I handed you a bag of sue-seki sword with a masamune in there and asked you which one was likely to be of best quality, you'd say the Masamune one and you'd be right ex-abrupto everytime. For most cases it's a statistical process with some probability of being right. I've seen these threads in the past go all the way down to "Everything lies in the eyes of the beholder/Quality and or Beauty has no basis in reality". Let's just say this would be a thread of its own more suited for a forum on the philosophy of aesthethics. There are many proxies for quality: Old records from acceptable gifts to Fujishiro/Meikan and various others, to Darcy's pass factor... Which is interesting to me. The fact that the fungability crosses through traditions and into a school. The big difficulty with judging through entire school is that some schools have very varied master-smiths, while for others there is an outlier who produces incredibly work and gets all the Tokuju while the rest are completely unknown. Shizu is a good example of this, so is the famous Bungo master. For Yamato this is made even more difficult because we simply don't know a lot about individual smiths given the paucity of extant work. Tegai Kanenaga and Shikkake Norinaga produced excellent blades and have probably become a way of saying top quality for the school more than anything else, but even this is problematic for me. Comparatively we know even less about Hosho and Taema lines. This is to say the fungability becomes extreme and judges are most likely loathe to stick their necks out and give a specific attributions at all. What has likely happened over time is that the few signed exemplars of the Yamato tradition ended up 'attracting' similar blades forming clusters of attributions based on quality. This leads us to clusters of blades attributed based on proximity with very few original and signed production. When you have so little data, you know nothing about the variance in the production of the smith compared to or those of another line. While classifications is expected to at least correlates with smith ground truth, it will be driven mostly by an appraisal of on the execution of some desirable trait (let's call that quality). So this is a hypothesis as to where we are with Yamato: clusters of blades judged based on some similarity, hovering around very few examples. I want to understand those clusters, when an attribution is made, what are the second and third guesses?

I like that he focuses on getting attractive packages of a sword and its Koshirae. Nick puts a big premium on cutting tests, interesting horimono, beautiful Koshirae and all sorts of things which add to the collectibility of a blade. He also offers great deals to members of this forum.

By T1 I mean 'Tiers' as in levels of quality. This isn't tied to the history of schools or makers as with sub-schools and offshoots, but rather it is a way to understand the process of attribution. What I mean by this is that attributions are a judgement of quality and this becomes more true the more fungible the categories for mumei blades. There are different paths blades can take based on their level of quality. For instance: Go T1->Tametsugu T2->Uda T3. If you get 'Den Tametsugu' it could be up on tier, meaning Go, or down one tier, meaning Uda. Another one is Norishige T1->Sanekage T2->Uda T3. Masamune->Shizu Kaneuji->Naoe Shizu. It also applies to other traditions, such as Awataguchi->Rai->Unju. Bizen as well, where the Soden-Bizen category gets quite flexible with Chogi and Kanemitsu on top, and the top students hovering nearby and Hasebe looming over the very soshu-like pieces. Basically if you think of attributions as a dart thrown at a graph for which nodes are makers (more precisely, one of the maker's style) and edges are 'paths' which can be taken for attribution when the work of different makers are correlated. There is embedded in this graph a hierarchy of quality, with T1 makers epitomizing the highest value. This brings us back to 'Den' which means you're not slam dunk on the node but your dart landed on the edge between two or more nods. The path of the dart has its own component of randomness in its flight path, which is the error in attribution and this is a way to think about blades that get upgraded from Juyo to Tokuju to a different maker. For instance, Tametsugu->Go. For Yamato, while the schools are very fungible, it's unclear to me how the graph is organized. I know Hosho sits on top, but what is sub-hosho? As in, Hosho but one level of quality less. How are the paths organized? Which school stand completely aside, and which ones are so close that its a tough call. Yamato is a big fungible cloud and I don't have its structure figured out right.

I'm still a confused with Yamato school. I know Hosho is T1, but I don't know the relative placements of the other schools, nor which ones are fungible or particularly distinct. I know Senjuin is recognizable because of the period shapes whereas some of the others are not. Is Tegai a way of saying sub-Hosho? or is it's own thing. I know these blades have a reputation of being very hard to judge and abundantly mumei, and thus there should be a lot of fungibility. I'd venture to say something like: Hosho T1->Taema T2->Tegai/Shikkake T3 and Senjuin being its own thing. But that's probably wrong.

This would need further studies and some hard number then. Maybe it's a simple chronological effect, with earlier sessions with more 'Den' compared to later sessions. That would be interesting and I wonder what it reflects: more conservative judgement, more certainty in judgement, submitted blades being more 'slamdunkish' compared to ambiguous early session submissions? (endogeneity?) Early Shinsa: Den Tier 1 Late Shinsa: T2 Or is it more something like: Early Shina: Den T1 Late Shinsa: T1 I think we all agree it's not: Early Shinsa: Den T2 Late Shinsa T1 Which would reflect the Den +5% Darcy in his blog post draws attention to the Muromachi cloud during the bad sessions. Which is clearly for session 23 and 24 (the worst sessions) but not 25. My second hypothesis as to where to look for drop in standards during these high volume sessions lies with the sheer number of sub-Hosho Yamato blades. We have a lot such blades in there volume wise. Some are certainly great and worthy, but one has to wonder what the distribution of quality looks like here. As for anecdotal value, I remember Fred sold 'Juyo in name only' Yamato Wak with a sessions number close to number 25 which we can refer as the cheapest Juyo ever offered on the web. We also have a quite a few suspicious Enju on the low-value side, especially the Enju Wak is a strange one.

Thank you Jussi. Once again, great and insightful work here. What strikes me is the number of 'Den' at Juyo-level attributions in Session 25. I don't have hard data to compare with, but this could be another sign of liberal attributions / maker inflation / drop in standards, on top of the sheer volume. No less than FOUR 'Den Tametsugu' and not a single Tametsugu slam dunk. This is a statistical anomaly, and what I take this 'Den Tametsugu' to mean is mid-grade nambokucho soshu work. A lot of 'Den Shizu' and even a 'Den Masamune'. 2/3 of the AOE attribution are also 'Den'. There is the +/- 10% interpretation of Den, and these blades could be on the minus side. We also know that archetypes slam dunk get a far easier time at Juyo. Strange.

I also enjoyed looking at this piece. I find the Koshirae very attractive. Not sure who exactly is the maker of the blade. Sendai Kunikune appears to be both a maker and a school.

I think we should qualify this statement further. Every sword from sources accessible to the vast majority of westerners has been assessed. Assessed doesn't necessarily mean Shinsa. When in doubt, Shinsa is a gamble and you may lose money at Hozon. It's also a time sink. Money now is worth more than money later. What it is guaranteed to mean is that someone professional in the food chain had a close look at it, and in most cases it was purchased in the sword dealer internal auction, meaning that a lot of experienced dealers gave their best guess. The big fish in Japan find new swords every year. In fact, one of them even has a business of handling estates with a group of experts, a glossy website, and a reputation. This is where new swords are found, which go on to fuel the shinsa process for the latest session. This is where the whales compete and profits are made: in access to these estates. Fresh, new, top blades are still found there, and judging by the inputs in Shinsa, it doesn't seem like the supply of top blade has dried out yet. I haven't seen any evidence of inflation, and in fact quite the opposite, which makes me suspicious of supply-side gaming a la DeBeers but this is a topic for its own post. Collector Estate->Big Fish->Internal dealer auctions for triage->Dealers->the vast majority of us. That's the food chain we're dealing with. Everyone at each level had a look, and applied his markup in order to make a profit, or not lose money at the very least. Sometimes mistakes are made, the lower the presumed level of the item and the more volume oriented the business model, the more likely the mistake because it's just not worth it to spend time in study.

I really enjoy your posts Jeremiah by the way, especially the examples you post here for examination. And I also think you have fine taste.

Take a Chinese sword and bake it in some rust-inducing chemical to create a fresh new lottery ticket. I find it amusing how one can basically discern a signature under the rust as if the fraudster left a piece of tap to keep the area from oxidizing. It's interesting to watch how the Ebay fraud scene constantly evolves and renews itself.

Dealers want to make money. If papering a sword increases the money they make, they will get the sword papered. This is simple. Those that do not act as profit-seeking go bankrupt and are replaced by dealers who follow market incentives. To say that dealers don't care about papers just does not make sense at all. If the blade has a big name, they absolutely care about getting Hozon. If the name is too small to even bother with papers, they don't care. This is why most iron tsubas and low grade nihonto aren't papered, it's not worth the cost and hassle. Dealers are rational creatures and those that aren't simply cease to exist. As for the general point, It's a question of probability. To say that all unpapared or old papered blades coming from Japan are gimei is obviously not true, just like saying that none of them are gimei. The real questions are, what are the factors that increase, or decrease, the probability of fraud? I use fraud here because Gimei is just one way. But passing off Shinshinto blades as Koto, or a mumei Hankei as a Shizu are all the sort of things which fall into this wider bucket of swindling. If it's unpapered with a big name, and sold by a dealer at a 'special price' then it's 99.99% gimei. If the dealer is online then it's even worst (less reputation consequences). If we go on Yahoo auction fishing for bargain Masamune or Kyomaro then we're deep into the fraud swamp. If it's sold by a widow's estate liquidator who happens to have found a few swords under the rug, one being signed Kanemitsu, then maybe it's Soshin. Chances to access the later for a westerner are extremely slim. This is how we should think, to break it down: 1. Expertise of the source. High expertise: Fraud %+ 2. Price. Price point below market to trigger greed: Fraud %+ 4. Name. Big name: Fraud %+ 3. Consequence of Fraud. Anonymous, low consequence interactions: Fraud %+ It's better to profess caution and skepticism than to let beginners get burned and drop out of the hobby.

We need to be realistic. This is what AOI believed when they put up the listing: The menuki are plated It's most likely gimei Now, AOI does make mistakes. It happens, such is the nature of high-volume businesses. My theory is that the previous owner, or AOI art, checked the Menuki and had an amateurish in shop Tsukamaki made and sold it off to the next guy as "maybe solid gold". This is how these things get cycled in the 'freak circus' arena of nihonto's ladder theory arbitrage. In this domain, the expertise and business acumen of AOI is simply unrivaled. It's possible to game AOI, you need to be lucky or incredibly knowledgeable and understand their business model. I would not recommend playing this game as it's like going one on one against the House, play enough times and you'll lose, and the beauty of it all is that you'll remain convinced you won. You might as well open up the lottery ticket. The tsuka-maki work is amateurish and possibly worth around ~50$. If these are solid gold Yokoya menuki, we're looking at 5k+ with very high upsides depending on the maker. Do I think it's the case? No, but sometimes the House makes a mistake. If you're confident its solid gold, then that's the rational thing to do. Making the case it's solid gold and referencing your expertise with precious metal on the board to give impressions that you're selling this at a 'discount' is the continuation of the AOI business model. Either you believe its gold and open it up, or you don't believe its solid gold and you sell it as is with your truthful impression.

To memory, this is the most desirable sword I've seen offered on this board. Signed Bizen Kodachi from a Jo-Jo saku maker, during a period for which it is hard to get Jo-Jo Saku as you get to compete with top Koto. Throwback shape with ko-kissaki gives you the elegance of olden times. Taken together, my view is that the ownership tax on this sword is low (you can sell close to what you paid) and the emotional dividends of ownership are high (it has the attributes of a valued blade). It's rare to see a sword like this circulating in the west. We are inundated by 'freak circus' blades and when a blade not only stands apart but shines like this one it deserves to be highlighted.

It has a good shape, classic archetype Nambokucho. Suriage is poorly done though. The coarseness of the hada throws me off. It does not scream Naoe Shizu to me at all, if anything I would have suggested Soden-Bizen. The blade is good but in my opinion the attribution is far from certain. I speculate it was tossed into the Naoe Shizu mega bucket at Juyo. That's an expensive, generally daimyo family habaki (although they can be sometimes used as 'dress up' so be skeptical). The sword was probably highly valued in the past, and perhaps thought to be of one of higher ranked maker. That's the kind of blade that would get a higher ranking soshu attribution during the great Honami inflationary attribution phase and hence could explain the habaki. Naoe Shizu is not a highly regarded attribution, and generally low-tier Juyo where 'cheap for Juyo' can really bite you hard. There is also quite a bit of variance in this attribution in my opinion and it requires a lot of knowledge to know differentiate the 'borderline Shizu' from the low Juyo ones. In my opinion, the one of Jussi is far closer to Shizu than the one linked on Aoi art. Finer hada, more typical hamon hint towards a tighter relationship to the master. That's classic Naoe Shizu. And nakago looks great. It's really a tricky school with a lot of potential for confusion, not least because its huge and spanned generations of evolving style, adding variance and creating a mega bucket. On top of it, for the non-experts, the distinctions between Shizu, Shizu Kaneuji, Yamato Shizu and Naoe Shizu is tricky business and means everything.

Those mounts will go Juyo. The maker is one of the best late Meiji makers. The dagger confuses me. 17th century does not seem correct, and the signature is not the typical EDO, deep chisel strokes but more a tentative old signature in the koto flavor. No idea what it is. Speculatively, and I can't read the Kanji, it could be one of the Shikkake line Norinagas, and signed work is incredibly rare. If that's the case, blade could go Juyo too. On the other hand, the Nakago seems to have a too 'thin' patina for this to be old Yamato work and the shaping of the Nakago is not good either, unless it's been reshaped. It's a bit of a mess and raises many red flags. My money is on some no name Edo smith. Original blade was high quality, and got swapped later. But the provenance is the original client for the Koshirae, which makes this unlikely. Puzzling. The mounts itself justify the price though. A typical case of Christies hiding a specialized item into Painting & Pottery.

A very fine Koshirae, and the blade appears promising. With better images of the blade we could be able to say more. This is worth a detailed look.

The real question we should pose ourselves is why is this even put into question. We do not need even to evaluate the workmanship, the signature, and these otherwise essential attributes but simply the characteristics of the exchange to infer whether or not it is Gimei. The following characteristics together ensures it is Gimei with a very high degree of certainty. 1) it is sold by a Japanese dealer (i.e has knowledge, as opposed to say a widow liquidating her husband's assets) 2) It has famous name and big signature 3) it has green papers This holds unless you believe that the dealer is irrational, or pro-social in that he desires to give you money. Ubu kanemitsu tachi in good health, signed and date is Tokuju/Bunkazai level. We're talking 100-300K USD. That's a very pro-social dealer. A word of caution: If you indeed know the dealer for fifteen years and you believe he has your best interest in mind, then assume it is misunderstanding: he assumes you know it is Gimei, but you like the sword. Sometimes in Japan things which are obviously needed in communication for westerners are left unsaid or merely left between the lines. It's easy for someone who has been dealing with swords for twenty years at a professional level to think that what is obvious to him is also obvious for you.

Makes perfect sense to me. Old times will cling to what they know, but the industry is changing and failing to adapt will assuredly become ruinous. "I like sword shows, therefore they are good and should continue as they are" is not an argument by any kind of measure. This is a real product/market optimization question. And as things stand now, consolidating all US-based shows into one major show, six month apart from DTI, stands out as an excellent idea. And get away from airport hotels. You want to get the thirty-somethings who have disposable income into the hobby. This means the you go for trendy and more expensive conference centers, which are better located. You want to follow a similar model as Art Basel, the upscale Museum + Gallery model. The grungy table layout, D&D convention in airport hotel style is just not going to kick it anymore.

Have to agree with Dwain. Remember, there is a lot of shill bidding going on there. Seller is anchoring price perceptions right there, hoping to get the future suckers to up their willingness to pay on future catches. It reminds me of pump and dumps, and other similar schemes. Brian, I think there are a lot more fake Tsuba with proper copper Sekigane than we all want to imagine. Think of this one as an outlier, where the sekigane wasn't properly redone in copper after the cast. This paints a grimly picture.

There is very little on Goto Seijo to be found outside Japanese reference work. Completely and utterly eclipsed by the grandiose heights of Goto Ichijo.

I would not buy it either and I completely agree with Geraint. I can add on the Nihonto side: I have no trust at all in that Nakago, it looks like a very fresh patina to my eyes. You clearly see yasurimine and there is active orange rust on it. I think it could be Meiji.

It is wise for you to ask this question. It is never just A or B. I come loaded with different advice: Don't buy waks, don't buy run off the mill Shinto katana, don't buy low-tier stuff in general. There is too little to learn and too much to regret. No, go get yourself exposed to top quality at a fraction of the price. 4500.- Go for a barbell strategy of extreme quality combined with an extreme flaw. Seek the most problematic top quality blade. The broken masterpiece. This will allow you to approach great quality work at a fraction of the price. Mumei Shinto or Shinshinto with Hozon attribution to Jo-Jo/Sai-jo Saku level smith is a good way to get quality at steep discount and these are in the 5K range (as opposed to 30K+ signed examples). Another path: Get a dead koto blade of a good school. Not dead because of over-polish: you want to see that Jihada. But dead because it has no Boshi, a Hagire, or because it was re-tempered. These are fatal flaws just like mumei shinto/shin-shinto will drop the price down to the ground and you will see beauty like you've never seen and learn like you never could. And how to orchestrate such a barbell strategy? The broken masterpieces aren't exactly common, unlike the Shinto Waks. Take opportunity: write a polite letter in English, introduce yourself, say what you're looking for. Get it translated by Markus Sesko in Japanese and checked for courtesy, and send it off to ~ ten dealers. Send a reminder after two months. Ask on NMB.

I also do not think there are any Oei-Bizen smiths beyond Jo-Saku, and just two smiths have any Juyo blades at all as far as I know from that time and place. Oei-Bizen Yasumitsu is perhaps on the underrated side, but it doesn't appear egregious to me like the case of Kencho. Oei gets quite close to the dark age of nihonto. Soshu dies during this time with smiths like Hiromasa, Bizen drops off abruptly, new powers settle in...such as the teacher of Muramasa who works during this time, I believe Jo-Jo Saku and probably the highest rated smith of that epoch. Perhaps from that time, the most underrated smith is Uda Kunifusa. He's one full saku above Chu-Jo Saku in my opinion. Those were dark, dark times...all the famous schools are dying or dead and would never come to be equaled again. Because Oei belongs to Koto, it is judge in relation to Koto. Whereas the Shinto smiths your describe are judged in relation to Shinto. Different periods, different baselines for these ratings. There weren't a lot of bright shining star during Shinto, so it's quite easy to take one of the 'Sai-jo crowns' there. Generalist artisans begin to be honored there, as with Myoju. Tadatsuna is evaluated with Hiromono as part of his skill factor. Kotetsu was an ex-armor maker and Hankei a gunship. Different times...

Alex, How to understand a relative rating system? Well, simply add baseline of skill from each period, which you add to the Fujishiro score. If we recode Fujishiro score from 1-5, and code period scores as follow: Kamakura +3, Nambokucho +2, Muromachi +0, Shinto +0, Shinshinto +1. Now if you take a Sai-jo saku Koto smith, his absolute rating is an '8' while a Saijo-saku Shinto smith is only a '5'. This is just for illustration as there are clear issues Fujishiro's relative classification system takes into account sub-periods it's not a true relative rating system in the mathematical sense of the word. If it were the case, we would say have 10 Sai-Jo saku smiths per periods and that's it. We have many more during Koto times than later times. This is not trivial because it means there is a relative component and an absolute component to model The baseline offset we devised is not constant per level of smith. The offset is more pronounced the higher the Fujishiro score. Kirill, I tend to agree with you w.r.t to the 'shadow effect' of being a 2+ gen. This is also where I see some undervaluation in the Japanese ratings. However, this is exactly the sort of 'error structure' you would expect if you devised a rating system. You need to start somewhere, with some sort of prior belief for the quality of work of someone without having access to much data, and if he comes second in the school's lineage, you'd be right more often than wrong by automatically assuming he's correlated but a notch under the founder which is in all likelihood a positive outlier. Thus, to move this 'prior' away from the baseline position that he's one notch below, you would need a lot of data and study.

This is a great example of a blade as a historical data point. The fact that it is signed AND atypical means a few things: If it were mumei, it would not get this attribution because it deviates from the smith's known work it's a very valuable datapoint to understand the range of the smith, and by extension his school and how it connects to later Soshu