publ.: November 2007
In the past twenty years, debate among historians and archaeologists of Japan appears to have shifted from whether Himiko and Yamatai should be located in the Kinai or the Kyūshū regions to how far backward traditional seriation-based chronologies in Yamato should be adjusted to accommodate Himiko’s florescence there (she died ca. 250 CE) as well as the early emergence of the Miwa court in the 3rd-century Nara basin.
Two recent books (Gina BARNES’ State Formation in Japan: Emergence of a 4th-Century Ruling Elite, and J. Edward KIDDER’s Himiko and Japan’s Elusive Kingdom of Yamatai: Archaeology, History and Mythology) represent opposing views in this newer debate centering on the mid-3rd-century divide separating Japan’s Yayoi and Kofun periods. BARNES’ work proceeds from the more conservative premise, probably shared by the plurality of Japanese archaeologists, that the immense Hashihaka kofun, representing a breakthrough in inter-regional cooperation, should be dated to around 270-290 CE. KIDDER’s book, in contrast, would prefer to see the Hashihaka mound – the early culmination of an ongoing process of tomb development and political centralization in the Yamato region – dated before 250.
The specific point at issue between BARNES and KIDDER is whether Hashihaka, the large, 280 meter long key-hole shaped tomb located in the southeast Nara basin, may possibly be the burial site of Himiko, the Japanese paramount of the early-to-mid 3rd century, known by that name only in contemporary documents written in the Chinese Kingdom of Wei. BARNES, suggesting a date of 280 CE for the construction of Hashihaka, notes that "there still remains the disjunction of up to 30 years between Himiko’s ostensible death date and the building of Hashihaka," although admitting that "we do not know if her tomb was raised immediately upon her death or took several years or decades to accomplish" (BARNES 2007:98). KIDDER, of course, will have none of this: "wherever Himiko was buried, most of the tomb should have been built during her lifetime," he states, flatly (KIDDER 2007:248). Although admitting that "Hashihaka . . . has been placed [by archaeologists] between 260 and 280," KIDDER also notes "it is a commonly held view that the earlier tombs [in the Nara basin] are too small to have been those of a local paramount. Only Hashihaka should be considered," pointing out that Japanese "archaeologists are constantly moving tombs to earlier years as dating techniques improve" (KIDDER 2007:253). This touches the core of KIDDER’s critique: the traditional dating methods that rely on pottery seriation Japanese archaeologists use to date Hashihaka to the late 3rd century do not provide valid results for the 3rd century mounded burials in the Yamato basin. The age estimates are too young.
KIDDER’s strongest evidence for this implied thesis are the dendrochronological dates calculated from cypress remains found in two burial mounds situated close to Hashihaka. These two mounds – located in the Makimuku archaeological area immediately to the north of Hashihaka – are Makimuku Ishizuka and Makimuku Katsuyama. In the 1990’s, a wooden board from Ishizuka was dated to 177 CE; in 2001, wood remains from Katsuyama were dated between 103 and 199 CE (KIDDER 2007:249, 247). Because these wooden pieces were identified by their excavators as used in the original mound constructions, they suggest building dates of circa 180 CE for Makimuku Ishizuka and 200 CE for Makimuku Katsuyama. According to KIDDER, ceramic finds associated with these wood pieces were, for Ishizuka and Katsuyama respectively, dated to Makimuku 3 New style and Furu 0. Following ISHINO (1992:191) and BARNES (2007:115, Table 5.3), the ceramic dating for both tombs can be placed in the early part of the second half of the 3rd century, approximately 250 to 275 CE, although KIDDER also admits that pottery as early as Makimuku 2 has been found at Katsuyama (KIDDER 2007:251). Measuring between the mid-point of the two dendroarchaeological dates from the Makimuku mounds and the Makimuku 3/Furu 0 date range suggests that the ceramic chronologies may be around 75 years out of synchronization with the more exact tree-ring dates. KIDDER (2007:248) predicts that "the relative dating system [of the ceramic typologies], which has never strayed far from its intuitional origins, will have to concede to the realities of scientific methods."
Table 1: Current 3rd-Century Nara Basin Chronologies
|dates CE||ceramic types||associated
|190-220||Makimuku 1||Shōnai 0|
|220-250||Makimuku 2||Shōnai 1||Katsuyama|
|250-275||Makimuku 3||Shōnai 2/3||Furu 0||Ishizuka?|
|275-300||Makimuku 4||Shōnai 3/5||Furu 1||Hashihaka|
Adapted from BARNES 2007:115, Table 5.3.
BARNES’ take on the dendrochrological evidence, of course, is rather different. She never mentions the CE 177 date from Ishizuka, reporting instead some 1994 tree-ring results from a log found in a peat mass in the Ishizuka moat that yielded dates between 315 and 325 CE (BARNES 2007:115). Yet as she points out,
This date range, however, in the early 4th century, only applies to the time of log felling and is thought to be much too late for the construction of the mound itself, considering the types of pottery associated with it. Ishizuka yielded Makimuku 1 type pottery, the earliest Haji transitional ware in Nara. The assignment of Makimuku 1 to the late 2nd and early 3rd centuries conforms with previous radiocarbon dates associated with Makimuku pottery, two of which were calibrated as prior to AD 240. (BARNES 2007:115-6)
Thus, at a single stroke, BARNES contradicts KIDDER’s assertion that Makimuku 3 New style pottery characterizes the Ishizuka mound, asserts that the pottery found at Ishizuka is consistent with a late 2nd century date, and implicitly emphasizes the possibly intrusive nature of wooden remains and – unless KIDDER is simply mistaken about the association of Makimuku 3 New type ceramics with the Ishizuka wood board – of pottery remains as well. The key to accurate chronological assessment on the basis of ceramic remains is not simply listing what remains were found but judging which pottery type most accurately reflects the period of a particular mound’s construction.
Does this mean, in effect, that KIDDER is cherry-picking his evidence? Yes, of course it does – for BARNES does the same thing! Whereas KIDDER ignores the 4th century log in his discussion of Ishizuka, BARNES ignores the more recent dendrochronological date of 177 CE from the very same mound. Neither presents a "balanced" account of "all the evidence." It is in the very nature of these chronological disputes that scholars with differing perspectives select out for special emphasis both the data and the lines of argument that are congenial to their particular points of view.
From my point of view, the most fruitful way of advance is to recast the whole problem and treat the dates associated with the Nara basin ceramic typologies as an assumption or hypothesis that requires continuous testing against independent evidence. There are now two such types of evidence: the dates derived from written historical evidence and those derived from tree-ring analysis. And both these types concur in suggesting that the hypothetical dates associated with 3rd century Nara basin ceramic types are about thirty years too high.
The historic, documentary evidence is well known: the Wei chronicles which testify that Himiko was buried in a mounded tomb "more than a hundred paces across," to adopt BARNES’ suggested translation (BARNES 2007:96). On the assumptions that Yamatai = Yamato and that Himiko died in 248 CE, the obvious candidate for this tomb site is the Hashihaka mound, the earliest such massive structure in the Nara basin. TERASAWA Kaoru, the archaeologist who excavated around the northern platform base of Hashihaka in the 1990’s, initially dated the ceramics he found there to Furu 0 (EDWARDS 1996:177). KIDDER admits that the most recent ceramic finds are no later than Furu 1 in type (KIDDER 2007:253). In the received chronology, Furu 0/1 concurs well with BARNES’ suggested dated of 280 for the construction of the Hashihaka kofun. But if we use Himiko’s death date as the criterion for dating this same kofun, the Furu0/Furu 1 horizon should be pushed back to about 250 or even very slightly before.
The dendroarchaeological dates derive from the studies cited by KIDDER, and are 177 CE for the Makimuki Ishizuka mound and 199 CE for the Katsuyama mound. If we accept BARNES’ contention that Ishizuka should be dated ceramically to Makimuku 1 and KIDDER’s earliest ceramic date for Katsuyama as Makimuku 2, we can estimate the horizon between Makimuki 1 and Makimuku 2 at 188 CE. The hypothesized date for this horizon in ISHINO’s Nara basin ceramic typology is 220 (ISHINO 1992:191), or 32 years later than the estimate thus derived from the dendroarchaeology.
These two independently sourced dendrochronological tests of the hypothesized Nara basin pottery ceramic chronology are in striking accord, and suggest that a better version of the chronology would shift all hypothesized dates backward by about 32 years; thus:
Table 2: Revised 3rd-Century Nara Basin Chronologies
|dates CE||ceramic types||associated
|158-188||Makimuku 1||Shōnai 0||Ishizuka|
|188-218||Makimuku 2||Shōnai 1||Katsuyama|
|218-243||Makimuku 3||Shōnai 2/3||Furu 0|
|243-268||Makimuku 4||Shōnai 3/5||Furu 1||Hashihaka|
Adapted from BARNES 2007:115, Table 5.3.
Of course, BARNES’ suggested date for Hashihaka would shift as well, from 280 to 248 CE.
Proposed dates associated with ceramic typologies are not self-verifying. With two different types of evidence – historic and dendrochronological – and three independent witnesses (the Wei chronicles, and the Makimuku Ishizuka and Katsuyama dates) all in concurrence, how many more witnesses will be needed before a trend in the evidence is clear?
BARNES, Gina L. 2007, State Formation in Japan: Emergence of a 4th-Century Ruling Elite, Durham East Asia Series 8. London & New York: Routledge.
EDWARDS, Walter 1996, In Pursuit of Himiko: Postwar Archaeology and the Location of Yamatai. In: Monumenta Nipponica, Vol. 51, No. 1, pp. 54-79.
ISHINO, Hironobu 1992, Rites and Rituals of the Kofun Period. In: Japanese Journal of Religious Studies, Vol. 19, No. 2-3. pp. 191-216.
KIDDER, J. Edward 2007, Himiko and Japan’s Elusive Chiefdom of Yamatai: Archaeology, History, and Mythology. Honolulu: University of Hawai’i.