Heading North and Offshore Again - Japan

The Japanese islands consist of the isolated Ryukyus in the south, then the central islands of Kyushu, Shikoku and Honshu, with Hokkaido to the north (Figure 3.5). So far, Japan has no convincing evidence for human occupation before about 40,000 years ago. If people crossed from Asia at that time they would have been dealing with sea levels perhaps 70-80 m below present, when Sakhalin and Hokkaido would have been joined together and to the Asian mainland. They remained joined until about

12,000 years ago. Honshu, Shikoku and Kyushu formed a single large island under low sea level conditions, separated by permanent sea gaps from Hokkaido to the north and Korea to the west, although the emergence of several small islands in

Korea Strait might have made the Korea to Kyushu crossing little more than a hop between visible land masses.

How did the first human settlers enter Japan? An initial arrival across sea is actually most likely because the oldest sites occur in Kyushu and Honshu, rather than Hokkaido. But from where? Three islands in the Ryukyu chain to the south, Ishigaki, Miyako, and Okinawa, have produced some very mysterious human and extinct deer remains from caves, with almost no associated archaeology, some dating greater than 30,000 years ago (Kaifu and Fujita 2012). To reach these islands would have required several sea crossings, some perhaps over 100 km in length since they lie off the continental shelf. A migration route from Taiwan, which was joined to China at periods of low sea level, with the northwards-flowing Kuroshio Current is therefore possible, but very hard to assess. Ryukyu in the Paleolithic remains a great mystery.

The best choice for an initial population homeland for the major Japanese islands of Kyushu and Honshu seems to be Korea. However, the very distinctive early industries of these two Japanese islands, with their edge-ground axes and transverse-bladed ('trapezoidal') points, are so far not directly paralleled in Korea at this time depth (circa

40,000 years ago) and seem also to be very rare in Hokkaido.24 This could reflect the great density of archaeological work in Kyushu and Honshu compared to the other regions, and both Korea and Japan certainly shared a rudimentary large blade technology at this time (Seong 2008). Possibly also, the first Japanese developed lithic industries suited specifically to the forested landscape of their main islands, given that the edge-ground axes would have been very useful for working wood (e. g., for boats). My understanding here is that this is the oldest edge-grinding technology in world prehistory Edge-ground pebble axes also occur after 35,000 years ago in Late Pleistocene contexts in northern Australia, but in Korea, China and Southeast Asia they are not yet definitely reported from such early contexts as in Japan. Obsidian was also collected for tools by these initial Japanese settlers from a small volcanic island called Kozu, in the Izu Islands, that 35,000 years ago would have been about 40 km offshore from the mouth of Tokyo Bay (Ikawa-Smith 2009).

Rather mysteriously, the edge-ground axes disappeared from the archaeological record in Kyushu and Honshu at the time of a major volcanic eruption, with widespread ash falls (the Aira-Tanazawa tephra), that occurred about 28,000 years ago. Whether there was a cause and effect sequence here is not clear, but the Japanese archaeological record after the eruption underwent major changes, initially with a short-lived introduction to Kyushu of a tanged point industry from Korea.25 Following this, there was a significant new immigration from eastern Siberia along the Sakhalin-Hokkaido-Kurile land bridge, but not into Honshu or Kyushu, where a different Jomon cultural tradition with early pottery developed after 20,000 years ago. The migration into Hokkaido took place a little before the last glacial maximum, prior to 24,000 years ago, and was associated with the introduction of bifacial points and microblades. Both of these now made their first appearances in Hokkaido as well as Korea (Aikens et al. 2009), with the microblades spreading onwards into the remainder of Japan (except Ryukyu), where bifaces already existed. Interestingly, this was also a period of major faunal immigration into Japan from several regions of mainland Asia, bringing in new

Species such as the wolf, brown bear, leopard, aurochs (wild cattle), and macaque monkeys (Van der Geer et al. 2010: 231).

Analysis of ancient DNA from skeletons excavated from several sites in Hokkaido reinforces the conclusion that these microblade-using immigrants were not the same population as the earlier makers of the edge-ground axes. Their descendants in Hokkaido continued the microblade tradition and shared a mtDNA ancestry with populations in the Amur valley in southeastern Siberia. A molecular clock traces this shared ancestry to about 22,000 years ago (Adachi et al. 2011), a date sufficiently close to the appearance of microblades in northern Japan (circa 24,000 years ago) to render these people too recent in time to have been the makers of the edge-ground axes. Instead, they were probably the main ancestors of the indigenous modern Ainu (nonJaponic-speaking) population of Hokkaido.25 The earlier edge-ground axe users in Kyushu and Honshu were perhaps ancestral to later Jomon populations there, and to some of the indigenous (non-Yayoi) genetic components in the modern Japanese population (Chapter 8).

Were humans in Japan before 40,000 years ago? This country today has huge numbers of excavations and archaeologists. A rather staggering 14,000 Paleolithic sites are believed to exist there (Izuho 2011), including at least 1792 microblade-bearing sites (Sato and Tsutsumi 2007: 53). Were Japan really to contain convincing Lower Paleolithic sites and archaic hominin fossils, we would surely have evidence by now, as we do in China from a less intensive archaeological coverage. This suggests that Japan really was settled first by modern humans about 40,000 years ago, with an industry of large blades, transverse-bladed points and edge-ground axes, introduced by a probable sea crossing from Korea. These industries were eventually replaced by various combinations of microblades and bifacial tools, the former introduced prior to 24,000 years ago via Hokkaido. The prehistory of Japan after the last glacial maximum will be discussed later, but given its equable maritime climate it is likely that these islands offered a favored location for human occupation in the late Pleistocene, operating as a kind of refuge from the northeast Asian continental cold.