No announcement yet.

Reading signatures

  • Filter
  • Time
  • Show
Clear All
new posts

  • Reading signatures


    Many people are put off trying to read Japanese signatures because they think it is too difficult, but in reality most are quite easy. What can sometimes be a problem is being able to find the characters in a dictionary but with a bit of practice even that gets easier. Sometimes characters were abbreviated when being written, whilst others are written a cursively. In my case I sometimes solve the problem but occasionally I am beaten and have to admit defeat.

    Let us start then by learning how to look up characters, or kanji.
    I use three dictionaries:
    Koop and Inada’s ‘Japanese Names and How to Read Them
    Nelson’s ‘Japanese ‘English Character Dictionary’
    James Self and Nobuko Hirose’s ‘Japanese Art Signatures’

    The last mentioned is not the easiest one in which to find the characters, but it list them in handwritten form, which correspond to how the appear in inscriptions ,and also gives the abbreviated and cursive forms of some of them as well. This can be very useful at times. The first two show the printed forms of the kanji, but Koop also contains a wealth of information on topics such as titles, family names, places and so on. If I cannot see the radical I use Koop (see why later) who also shows some old fashioned characters that are not in Nelson.

    The first thing to note is that the Japanese copied the kanji from the Chinese in the first millennium of our era, applying the Japanese sound or pronunciation to their meaning. Because Japanese is a totally different language to Chinese, each kanji has therefore two different possible readings or pronunciations: how it is sounded in Japanese – the kun reading, and how the Japanese thought the Chinese pronounced it – the on reading. Many on readings have gone out of use in Japan and are always pronounced in the kun way, other kanji retain both. Which reading a Japanese would use to pronounce a kanji is really a matter of usage, although in signatures it will be generally be the kun reading. In all cases the meaning is the same for either reading, but be warned some kanji can have a bewildering number of meanings and which one is what is meant can only be decided by the context.

    As an example of all this let us take the kanji . The on reading of this character can be MEI, MIO or MIN whereas the kun reading can be AKI or occasionally AKIRA. The meaning of this kanji is ‘to dawn, begin, unlock, open, reveal or evacuate’. All this looks confusing but it isn’t really since in an inscription it will be obvious what reading or meaning to choose.


    Now let us look at how to find a character in the dictionaries. There are two features of a kanji that are used to identify it: the stroke count and the radical. First the stroke count. This is simply the individual marks made by the brush or chisel when writing the character, including any dots or ticks. The only slight complication is that a horizontal drawn from left to right that then turns downwards is considered as one stroke since the brush does not need to be lifted off the paper. So, 支shi ‘a branch’ is counted as 4 stokes since the lower of the two horizontals then turns down. What then is a radical? It some cases it can be the whole kanji but more often it is a part of one that indicates what the kanji refers to. Here are a couple of examples:

    木 this character ki means ‘a tree’ and is a kanji in its own right, but is also a radical that forms part of characters that refer to trees, timber, woodworking and so on. Thus matsu is a ‘pine tree’ and sugi is a ‘cedar tree’. This particular radical is generally on the left of a more complex kanji but it can also be in another position such as ka meaning a ‘frame, stand, shelf’ and so on. If you see a character with the ‘ki’ radical forming a part of it you know it is probably something to do with wood or trees or similar and it will be located under that radical in the dictionary. Similarly the radical is read kin and on its own means ‘gold or metal’. Again, when it occurs in a more complex character it means that character is something to do with metals or metalwork. Thus tetsu’is ‘iron’ whilst sabi is ‘rust or rusty’.

    Most dictionaries contain a table of radicals in the order of the number of strokes in them, listing all the characters using that radical together. Thus in Nelson is radical 167, which starts on page 904 and continues to page 918, listing all the characters in the order of the number of strokes in addition to the radical. Thus tetsuwill be in that part of section 167 in which characters with 5 extra strokes are listed whilst will be listed further on in section 167 because it has 8 extra strokes. Koop and Inada is essentially the same except the kanji are classified under the total stroke count first, including those in the radical. Thushas a total of 13 strokes and will be listed in that section, only then being divided into those with the same radicals. Both methods are really the same, just done in a different sequence.

    So, if we take the kanjitetsu, in Koop as an example, we would first turn to characters with 13 strokes, then go through that section of the dictionary to find all the kanji with an 8 stroke radicals, the kin radical characters being grouped together. In Nelson we would look up the 8 stroke radicals inside the front cover and find that kin is radical number 167. We would then go to section 167 and look at kanji with that radical that have an additional 5 strokes.
    __________________________________________________ ________________________________ MAKER’S NAME

    Now that we know how to find the characters in a signature, we can turn to actual signature and how to translate them. First it is important to realise that a craftsman or artist usually signed with a name called a nanori or ‘signature name’. This was not his or her real personal name, it is more of what we would call a ‘nom de plume’ or the name they are known by outside the family. These nanori are almost all made up of two kanji and most often read with their kun readings.

    Examples would be: 正光 Masamitsu,  兼道 Kanemichi 国定 Kunisada

    If something is signed with two characters, that is all there is to it. It is the name of the maker. Quite often the name will be followed by a character(s) that indicates the named person actually made the item. There are three common ways this was done:

    saku meaning ‘made’, 作之 saku kore meaning ‘made this’, tsukuru meaning the same sort of thing.

    So, 国定作 reads Kunisada saku or ‘Kunisade made (‘this’ being understood but not written)’
    兼道作之 Kanemichi saku kore or ‘Kanemichi made this’

    __________________________________________________ ________________________________ PROVINCE AND ADDRESS

    The next thing most commonly found in signatures would be the maker’s place of work, or address. Remember that for most of Japan’s history the lower classes were not allowed surnames and often used the place in which they lived as a substitute. They may however claim decent from one of the more ancient clans, adding the clan name before their nanori. In addition, during the Edo period these things were less strictly enforced and some craftsmen adopted surnames - more of which later.

    Feudal Japan was divided into a number of ‘countries’ or kuni for administrative purposes (note: this kanji is in a modern font and in old signatures it appear a bit different, but always a box, generally with a diagonal line and other strokes in it). We refer to these ‘countries’ as provinces.

    Each kuni has a two character name that is read in the kun manner and generally followed by the kanji for kuni. Thus:
    備前国 Bizen kuni, 山城国 Yamashiro kuni, 出羽国 Dewa kuni

    In the same way that we abbreviate the names of our counties, such as Lancs for Lancashire or Yorks for Yorkshire, the Japanese abbreviated their province names by taking one of the kanji in the name, often the first but not always, and adding shu.

    Thus: 備州 Bishu, but note: 城州 Joshu and Ushu

    In a signature the name of the province will normally be followed by ju meaning ‘living at’ or 住人 ju nin meaning ‘inhabitant of’. Let us now construct a typical signature with the components we have dealt with so far:
    備前国住正光作 which reads ‘Masamitsu living in Bizen province made (this)’

    The kanji in red can be regarded as ‘pointers’; the first ju tells us all to the left will be an address and that the characters before kuni are a province name (in reality it will be the part above since it would be written as a vertical column read from top to bottom), and the two characters above saku will be the maker’s name. So by remembering these two important kanji, you know what the parts of the inscription are even if you cannot read them without looking them up. If the signature also mentions a town or village, it will be positioned after the province name and before the ju character. Thus a famous sword making centre was Osafune in Bizen province. This would be written as:
    備前国長船住正光作Bizen kuni Osafune ju Masamitsu saku
    Again, you may not know how the characters before ju are read, but you do know it is an address.
    __________________________________________________ ________________________________


    Now above I mentioned clan names and family names. Claims that a person was descended from one of the ancient clans is common and since there are very few of them, they are easy to spot. Only about 3 occur regularly and of those the first two are by far the most common.
    These are: 藤原 Fujiwara  Minamoto  Ki 
    These will be positioned after ju and before the name:
    備前国長船住藤原正光作 Bizen kuni Osafune ju Fujiwara Masamitsu saku

    If the maker uses a surname it will be in the same place but obviously not one of the three clan names listed and will need to be looked up in the dictionary. Koop would be the best one in this case as he lists many family names in text under the first character.


    During the Edo period in particular, it was customary for swordsmiths to be granted honorary titles. These take the form that implies a high ranking position within a province, always one different from that in which the smith worked. Two ranks were commonly awarded: Daijo or Kami, the latter being the higher rank and approximating to ‘Governor’. Again, when present the title follows the address in the inscription and proceeds the clan name. Thus:

    備前国長船住山城国守 藤原正光作 So now we have Fujiwara Masamitsu declaring that he has been awarded the title of Governor of Yamashiro province despite the fact that he actually lived in the village of Osafune in Bizen province. The character kami is another pointer that indicates the province preceding it isn’t the address but a title.

    Reading dates

    The Japanese recorded the passage of time in several ways, the most common being the use of periods of time measured in years called nengo. Each nengo has a two character name pronounced in the kun reading and had a duration that could be a short as a year to many years. In the past new nengo were started for all sorts of reasons such as a violent earthquake or the birth of a new prince. Today, the nengo is changed only on the death of an emperor. Because the Japanese calendar was based on the moon, not on the sun as we do, a year designated in a nengo might differ by a year from how we would express it but the difference in dating will be of little consequence. What do not correspond to our calendar are the months and days. These require complex astronomical tables to determine their equivalence.

    A typical date would look like this:

    明治十二 The characters in red are the three key characters. These are nen, year, gatsu, month, and nichi or day respectively. The name of the nengo are the two characters at the beginning; in this case Meiji. The nengo Meiji started in 1868 and the numerals following that name state it is the 12th year of that period. Now, since the first year is 1868 the twelth year is 1868 + 12 - 1 = 1879. Following the character nen for year is the number 5 so the date is the fifth month in 1879, not May in our calendar but the fifth Japanese month after the start of their lunar year.

    For reference, the numbers are:
    ,,,,,,,,, 14= , 40=四十, 44=四十四

    An alternate dating system borrowed from the Chinese involves a repeating zodiacal system of 60 years. Thus a date might indicate 1560 or 1620 or 1680 and so on. This is often added to the nengo system after the nen character, and generally offset to the side of the column, just to make sure. This system involves combinations of two characters, one representing the 12 animals of the zodiac and other characters representing ‘stems’. Since this system is only very rarely used on its own it will not be dealt with here.
    Ian Bottomley - UK

  • #2
    Thank you Ian,

    this helps me try to solve the riddle of reading signatures that haunts me since 30 years ( I tried to solve it by convincing my eldest daugher to study japanese at the University: It cost me a lot of money, she graduated last year, monday she is leaving to stay three months in Osaka -her fourth trip to Japan- but she still has to translate me a single mei !)


    • #3
      Outstanding writeup, Ian! Took awhile to get through

      Massimo, you must demand at least three translated signatures/month.
      Return on investment, you know

      Jan - Sweden


      • #4
        Very nice complete explanation about signed items from Ian..
        Very important to translate a signed item and not always easy sometimes hours to understand clearly the kanji and could be erased or damaged with rust on iron for exemple ...
        this post and other Ian post about dating are really complementary and necessary when someone begin to collect or would understand some japanese art signed items


        • #5
          The following should be carved in stone, as it is the most lucid and down to earth explanation I've come across:

          "The first thing to note is that the Japanese copied the kanji from the Chinese in the first millennium of our era, applying the Japanese sound or pronunciation to their meaning.

          Because Japanese is a totally different language to Chinese, each kanji has therefore two different possible readings or pronunciations: how it is sounded in Japanese – the kun reading, and how the Japanese thought the Chinese pronounced it – the on reading.

          Many on readings have gone out of use in Japan and are always pronounced in the kun way, other kanji retain both.

          Which reading a Japanese would use to pronounce a kanji is really a matter of usage, although in signatures it will be generally be the kun reading.

          In all cases the meaning is the same for either reading, but be warned some kanji can have a bewildering number of meanings and which one is what is meant can only be decided by the context."

          But then that's the Yorkshire gene kicking in........

          As any Fule Kno.

          Pip Pip Cheerio



          • #6
            An excellent writeup! I've already learned many new things just from this


            • #7
              Laurent, The problem of reading signatures that is rusty, or damaged in some way, can be a problem. The first obvious thing to do is to take off the secondary rust using our well tried oil and bone scraper method. That is sometimes all that is needed but more often the strokes of the signature will still be filled with rust. What I then do is go at it with a needle, picking out the rust from what I can see. I am lucky in having an ancient binocular microscope on a long arm that clamps to the table which often reveals a lot more detail. You can also try angled light from different directions. Doing this will sometimes give you enough parts of the kanji to work out what it is. I have a sword I bought when I was about 15 with a very corroded tang. I could see there had been a signature but only one kanji, HIRO, was really visible gradually I have worked on it over the years and finally read the next kanji MITSU. I could see parts of kanji above the name and have then worked through the HIROMITSU smiths and have an idea of the province. Difficult but it helped.
              Ian B
              Ian Bottomley - UK


              • #8
                Ian , yes i confirm you re lucky with a binocular microscope it must be really easier ..what size /growth the microscope you use ?
                me first i put oil wait a little and sponge the overflow and some rust possible ..sometime i do it twice in order to watch ,after i take photos outside with outdoor light and photos with flash inside house and sometime i confirm with a small loop..
                a real treasure hunt isn t it ?!


                • #9
                  Dear members! What about little fun in reading the signature on my kabuto. The kabuto is made of nerigawa.

                  Bartek Bartosz Pasternak - Poland


                  • #10
                    Ok Bartek, I'll rise to the challenge and start it off with:

                    安 An
                    永 Ei
                    四 Yon

                    (An Ei era 4th year - 1775)

                    I now step down for another Katchbunny to have a crack.
                    Last edited by Malcolm; 04-02-2018, 02:46 PM.
                    Pip Pip Cheerio



                    • #11
                      Excellent Malcolm! It is 1775 indeed.
                      Bartek Bartosz Pasternak - Poland


                      • Malcolm
                        Malcolm commented
                        Editing a comment
                        Cheers Bartek!!

                    • #12
                      OK, I will now add that the smith / kawashi used the name 'TANAKA ...' - now over to someone else.
                      Ian Bottomley
                      Ian Bottomley - UK


                      • #13
                        Great Ian! Tanaka is the right answer. One more clue to solve.
                        Bartek Bartosz Pasternak - Poland


                        • #14
                          The last part of the clue: Gensuke.
                          Sadly I haven't found any clear reference to Gensuke Tanaka. However, there was Gensuke Iwai in early Edo period according to Japanese Armour Makers for The Samurai and in Shin Katchushi Meikan Gensuke is reffered as Sadakatsu who eventually changed his name to Gensuke Sadakatusu.

                          Is it possible that Gensuke Tanaka was associated with Iwai school?
                          Last edited by Bartek; 04-05-2018, 06:34 AM.
                          Bartek Bartosz Pasternak - Poland


                          • #15
                            It's an IWAI kabuto that was assembled for the Kuroda Clan.
                            I have one that is in the same style, nerigawa.

                            Note: Shikoro the Chikiri Odoshi, Yahadzu Kashira Itamono Zane.
                            Maki-e kin fukurin.

                            My bucket.....
                            Mei inside Kabuto in shu-urushi


                            岩井 Iwai

                            源之丞 gennojo
                            勝 katsu

                            Chappelear - Japanese armor makers for the samurai
Iwai Gennojo

                            Mounted armour for the Kuroda Jo I Chikuzen
Mid Edo Period

                            Sasama - Katchu Shi Meikan
Tsuguie (次家)
Iwai Tsuguie (岩井次家), early Edo period, lived in Fukuoka (福岡) in the Sagara district (早良郡) of Chikuzen province.
                            He was the son of Iwai Ienaga (家長) and an armourer of the Fukuoka fief which was ruled by the Kuroda family (黒田). His first name was „Gennojô“ (源之丞). According to the „Chikuzen zoku-fudoki“ (筑前続風土記), Ienaga was quite a famous master so we can assume that the same applies to Gennojô Tsuguie too.

                            IMG_7493 by Katchushi Koubou UK David Thatcher, on Flickr

                            IMG_7498 by Katchushi Koubou UK David Thatcher, on Flickr
                            Last edited by DaveT; 04-04-2018, 06:43 PM.
                            David Thatcher
                            Professional Armour Restoration (Katchushi Koubou)
                            Facebook Group: