If you work with Japanese picture postcards as historical source material, you have probably noticed that most surviving cards lack postmarks and other temporal indicators. This post provides a guide to estimating the age of these revealing artifacts, based on printing conventions that conformed to international and Japanese postal regulations. This post also provides several examples and explanations of Japanese postmarks.
This chart by Urakawa Kazuya provides the basic method scholars use to subdivide Japanese picture postcards into four periods:
Period I. Undivided back: October 1900-March 1907
Period II. 1/3 divided back: March 1907-March 1918
Period III. 1/2 divided back: 郵便はかき: March 1918-February 1933
Period IV. 1/2 divided back: 郵便はがき: February 1933-August 1945
Urakawa Kazuya 浦川和也. “Kindai Nihonjin no Higashi Ajia, Nan’yō shotō e no ‘manazashi’: ehagaki no rekishiteki kachi no “ibunka” hyōshō” [The Japanese “Gaze” on the Peoples of East Asia and Micronesia: Archives Importance and the Other Race Representation, in the Japanese Picture Postcards]. Kokuritsu rekishi minzoku hakubutsukan kenkyū hōkoku 140 (March 2008): 133.
One side of the postcard, the front, is dominated by a photograph, design, painting, drawing, or image. The back is reserved for addresses, and after 1907, for addresses and messages. In Japan, picture postcard history begins in 1900. Before this time, there were postcards, but due to postal regulations, it was not possible for companies to produce picture postcards until 1900.
The back of the card lacks a dividing line to separate the message and the address.
1/3 Divided Back
The back is divided so that 1/3 of the space is reserved for a message, and 2/3 is reserved for the address.
1/2 Divided Back
The back is divided so that 1/2 of the space is reserved for a message, and 1/2 is reserved for the address.
East Asian Image Collection Periodization
Based on research into the postmarks and cards themselves, contemporary journalism, secondary scholarship, and some postcard websites, the EAIC has devised the following periodization scheme:
Period I. Undivided back: October 1, 1900-March 27th, 1907
Period II. 1/3 divided back: March 28, 1907-February 28, 1918
Period III. March 1st 1918-February 14, 1933:
1/2 divided back w/ 郵便はかき OR
1/2 divided back w/ 郵便ハガキ (note: the katakana usage changes the ka/ga rule)
Period IV. 1/2 divided back w/ 郵便はがき: February 15, 1933-August 15, 1945
Discussion and Illustrations
On October 1st, 1900, it became legal in Japan for private companies and individuals to produce picture postcards that the Japanese postal authorities would accept and deliver at the postcard rate. Precise regulations were published regarding the size of the postcards, and the format.
During this earliest period for picture postcards, the only information allowed on the back of the postcard was the addresses of the sender and and the recipient. Any messages needed to be written on the front, or picture side, of the card. Here is a postcard stamped with the date “December 27th, 1905.” It was sent from a Japanese colonial official in Taipei to Nara Prefecture. Notice that Japan is called “Yamato Kuni” here.
Here is the front of the postcard, which has both the picture and the message from the sender (this is a new year’s card for 1906).
Here are some more examples of undivided back cards from Period I:
Some undivided back cards are military mail from the 1930s and 1940s. It should be clear from the type of cardstock, imagery, and other clues that such cards are not from Period I:
On March 27th, 1907, new postal regulations made it possible to use 1/3 of the back of a picture postcard for written messages. The card below was issued on June 17th, 1906, to commemorate the eleventh anniversary of the Taiwan Government General. Accordingly, its back is undivided, since this date falls into Period I.
The next card was issued a year later to commemorate the twelfth anniversary of the Taiwan Government General, on June 17th, 1907. Since this date falls into Period II, the card has a 1/3 divided back:
Here are some other examples of 1/3 divided back cards:
On all prewar (1900-1945) postcards, the phrase 郵便はがき is written from right-to-left, as きがは便郵. When hiragana was used, pre-1933 cards omitted the “ten-ten mark” that turns “ka” into “ga” (see card on right). When katakana was used, however, the “ten-ten” was retained (see card on left):
From March 1st, 1918 through February 14th, 1933, Japanese picture postcards were printed with 1/2 divided backs, with the “ten-ten” in が (ga) omitted:
This next card retains the “ten-ten” in “ga.” Nonetheless, it falls into Period III because katakana is used to write out “hagaki.”
Because the front of this card is clearly postmarked “October 30, 1928,” we know it could not have been published in Period IV.
The EAIC contains 71 postcards with katakana “郵便ハガキ” printed on 1/2 divided backs. None of these cards are demonstrably post-1933. Many of them are indeed demonstrably from the 1918-1933 period. Based on this evidence, I classify 1/2 divided back cards that have “hagaki” written in katakana as Period III cards.
After February 15, 1933, the “ten-ten” was added to the “郵便はがき” marking on the backs of Japanese picture postcards. Here’s a postcard published in 1932, before the change in postcard printing conventions, but after the “Manchurian Incident”:
This postcard was published in or after February 1933, which is evident from the “ten-ten” on the kana for “ga” in “hagaki”:
This “1/2 divided back, hagaki” style card was published until the end of the war (August, 1945)
After the war, “きがは便郵” was reversed to read “郵便はがき” as shown in this next card from the American Occupation Period:
Some Japanese picture postcards from Periods III and IV do not have the standard 1/2 divided back “hagaki” mark. These exceptional cards are military mail, “real photo” postcards, and postcards produced for foreign consumption–as far as I can tell.
1926 Taipei Industrial Exposition Real Photo Postcard
Commemorating an August 1933 Air Defense Drill
Commemorating invasion of Shanghai, 1937
Manchukuo and China (郵政明信片)
Japanese picture postcards published for use in the postal systems of “Manchukuo” (1932-1945) and “The Chinese Republic” (1940-45) employed Chinese-language terms for “postcard” (明信片) and “post office” (郵政). Both countries were nominally sovereign and yet were governed by Japanese military and civilian officials. Thus, it is not surprising to find that each nation, no matter how “fake” in retrospect, possessed distinctive flags, postage stamps, currency, and postal regulations. I am still trying to work out a periodization scheme for these cards. What follows are some examples of Chinese-language markings on Japanese picture postcards.
This first example is postmarked “August 19, 1932.” The lack of a “ten-ten” on the “ga” indicates that it was issued before February of 1933. The front depicts a scene from Harbin, which had a significant Russian population in the early 1930s. The front of the card is bilingual Russia-Japanese. The back says “postcard” in Russian, Chinese, and Japanese. It was mailed by military packet, as indicated by the red “軍事郵便” stamp. My guess is that this card was designed for sale to Russian and Chinese residents of Harbin, as well as Japanese soldiers, settlers, and officials. The back of the card suggests that it was intended for use in the postal systems of all three countries.
Here’s another card from the same period, but with “満洲國郵政明信片” “Manchukuo Postal Service Postcard.” The Chinese-language text on the back now includes instructions in the stamp box to affix a stamp, and instructions to reserve the right half of the postcard for the address (along the dividing line). The Japanese word “郵便はがき” is retained, and the card was produced by Taisho, a Wakayama Prefecture based company. So this is Japanese postcard that was produced, it would seem, for use in the Manchukuo postal service.
According to Shimada Kenzō’s guide to Japanese commemorative postcards, the Manchukuo Postal authority was established on March 9, 1932, under the administration of the Transportation section. On April 4th, the Manchukuo state seized control of Republican China’s Liaoning and Jilin postal bureaus, but the International Postal Union refused to recognize this gambit on April 24th, 1932. In July, the Nanjing government refused to handle Manchukuo postage, withdrew their postal workers, and cut off the northeast provinces. In response, the Manchukuo government took over all Chinese mail service, and began to issue its own regular stamps and postcards on August 1st, 1932 (Shimada 2009, p. 214). Daqing Yang notes that the Republican Government was unable to isolate Manchukuo in terms of international postage because the Japanese postal service could route mail through the Kwangtung Leased Territories (Yang 2010, p. 81).
The well illustrated and documented website “Manchukuo Stamps” provides examples of government issued postcard backs that resemble the card under discussion: these were first issued on 26 July 1932. Thus, it would seem that this card, and others like it, were issued between July 26, 1932, and February 14, 1933, since all seven examples that I’ve looked at have a か instead of a が for 郵便はがき.
On March 1st, 1934, Pu Yi was declared the “Kangde Emperor” and Manchukuo was recognized as an “Empire” by the Japanese government. The new postal service now appeared as 満洲帝國郵政 on government-issued “imprinted” postcards. The Manchukuo Stamps site shows this new imprint in effect by November 1st, 1934. Consistent with the new status of Manchukuo as an empire, the phrasing 満洲國郵政 should have fallen out of use after March 1, 1934. Tentatively, I would estimate the date of postcards with “満洲國郵政” on the backs as falling between July 26, 1932 and March 1st, 1934.
This next example is from Mukden/Shenyang at the Royal Tombs of the Manchu Emperors. There is no “郵便はがき” mark, but only “郵政明信片.” It was published by the “Shinkyo Tourist Society.” Shinkyo 新京 was Japan’s capital of Manchukuo (the city of Changchun).
This new generic Chinese word for postcard, without a trace of “Manchukuo,” was put on government-issue imprinted postcards from Manchukuo from January 1935 (Manchukuo Stamps). After the Republican government in Nanjing attempted blockade of Manchurian mail, the Japanese pressed for a renewal and strengthening of communication ties (Yang 2010, 99-100). An agreement was reached in December, 1934, to lift the ban an Manchukuo mail in Republican China. Thereafter, 郵政明信片 was used on Japanese official imprinted postcards from Manchukuo into the 1940s. Thus, I would tentatively date Japanese postcards with 郵政明信片 markings as dating from January, 1935 through the early 1940s (I do not know the end date).
It should be noted that the overwhelming majority of Japanese postcards with Manchurian themes retain the postal conventions outlined in the Periods III and IV section above (1/2 divided backs with 郵便えはがき marks in Japanese.
This next card is marked “Republic of China” 中華民國郵政. These are Japanese postcards designed for use in the Chinese postal system. This marking can be found on postcards dating as early as June 10, 1932, about six weeks after the Chinese Postal Bureau was unified under a single organ on May 1, 1932. Until I can find counter-examples, I assume cards marked like the one below were published after 05-01-1932.
There is a fascinating history of Sino-Japanese friction, diplomacy and intercultural relations to be disentangled with reference to these above mentioned postcard conventions. Kishi Toshihiko 貴志俊彦 has discussed these issues in his Manshūkoku no bijuaru-media: Postā, ehagaki, kitte 満洲国のビジュアル・メディア: ポスター・絵はがき・切手 (Colonial Manchuria’s visual media: Posters, pictorial postcards, postal stamps). I currently lack sufficient data to propose a periodization scheme for these types of cards.
Postmarks and Dates
Japanese Calendar Postmarks
The most common postmarks on prewar Japanese picture postcards look like this:
The basic rule is: [Japanese Emperor Reign] Year-Month-Day.
The numbers represent reign years of the Japanese Emperors. These reign-year/Gregorian calendar conversions are relevant to prewar Japanese picture postcards (1900-1945):
Meiji Period=Year 33 (1900) through Year 45 (1912)
Taisho Period=Year 1 (1912) through Year 15 (1926)
Showa Period=Year 1 (1926) through Year 20 (1945)
Using the rules established in above section for Periods I-IV, it is usually clear which reign period a postcard falls into. For example, year “10” cannot be Meiji 10 (1877), because there were no picture postcards in 1877. If the postcard is Period III (1918-1933) [it has a “ka” instead of a “ga” in “hagaki”], then “10” means “Taisho 10” or 1921. However, if the card is Period IV (1933-1945), then it is “Showa 10” or 1935.
The number “7” in the year column would make a card either 1918 (Taisho 7) or 1932 (Showa 7), both of which fall into Period III. In general, postcards from 1918 and 1932 are sufficiently different in terms of lithographic technology, thematic emphasis, and other factors to make a determination possible.
Gregorian Calendar Postmarks
Much less common are postmarks with Romanized place-names. The rule for these stamps is: Day-Month-[Gregorian Calender] Year.
On this card, “朝鮮” is written “CHOSEN.”
The written message is dated “Nov. 2,” so we can infer that the “3.11” refers to 3 November, and that the card is postmarked on November 3, 1923. This date is consistent with the Period III markings on the rest of the card.
This card has two examples of romanized place-name postmarks:
Following the rules outlined above, we ascertain that the letter was signed by a letter-writer named Nobuyuki in Busan [“Fusan”], Korea on December 10th, 1911. The card arrived in Shimonoseki, Japan on December 11th, and in Kobe, Japan on December 12th, and probably landed in New Bedford, MA, USA on January 11th, 1912. The faint mark of “Meiji 44” is visible in the postmark near the stamps box, revealing that a “normal” postmark was also affixed to this postcard simultaneously with these more “foreign” styles.
Japanese Calendar with Sino-Japanese Numerals
During the first decade of the twentieth century, some Japanese postmarks took this form:
Sometimes, these were the only postmarks on the card:
Other cards have both types of postmarks. On the card below, the Sino-Japanese numeral postmark was stamped twelve days earlier than the Hindu-Arabic numeral postmark.
Sexegenary (60-year cycle) Years
Lastly, Japanese postcards passing through the Republic of China (中華民國) postal system were sometimes marked with sixty-year cycle years. The following card was postmarked in on May 26th, 1914:
Here’s the “Republic of China” postmark (中華民國)
The date portion of the postmark reads: 甲寅 (blue) 五月(red) 廿六 (yellow)
The year “甲寅” is a 干支 or “Stem-Branch” or “sexegenary-cycle” year name, which has a very ancient pedigree in China, Japan, Korea and other places. This chart assigns Gregorian Calendar years to “stem-branch” years:
The above guide is subject to revision as more data becomes available and my research efforts deepen. As far as I know, there is nothing controversial about the dating methods proposed in the foregoing. This research is made possible by Eric Luhrs, James Griffin III, and Paul Miller at Lafayette College Digital Scholarship Services. This team has been designing software, capturing images, producing pdf files, and otherwise putting thousands of images at my fingertips, making it possible to connect these various types of evidence in a meaningful way.
Sino-Japanese Calendars and Dating Systems:
For an accessible but thorough introduction to reign years and the stem-branch cycle, I recommend this blog post by “Kristen D.”
For a comprehensive set of tables to align the Korean, Japanese, and Chinese calendars with the Gregorian Calendar, back to ancient times, the following is indispensable for scholars who wish to date artifacts in East Asian History: