日本舊石器晚期石器技術(shù)起源的新考古學(xué)與人類學(xué)證據(jù)

摘要：對日本舊石器時代晚期開端的探討主要通過兩種模式展開，即列島內(nèi)舊石器時代中期演化模式和大陸舊石器時代晚期擴散或遷徙模式，然而，最近來自日本和周邊國家的考古學(xué)證據(jù)正對這種簡單的模式提出挑戰(zhàn)。本文批判性地回顧了包括可能存在的早中期在內(nèi)的日本舊石器時代已有年代序列，并嘗試展示有關(guān)日本舊石器時代晚期開端的另一種模式。本文列舉了一些可能屬于舊石器時代早中期的發(fā)現(xiàn)，并建議進一步開展地質(zhì)考古研究，以了解其可信度以及與舊石器時代晚期早段（EUP）之間的文化關(guān)系。目前，日本EUP 的開端以距今約3.9-3.7 萬年前古本州島具有臺形石器和鋸齒刃器的石片工業(yè)為特征。盡管臺形石器僅流行于日本EUP 階段，并且可能源自島內(nèi)更早的石器傳統(tǒng)，但這一石器工業(yè)也在一定程度上顯示出與同時期中國和朝鮮半島石器組合的相似之處。石葉技術(shù)最早出現(xiàn)于古本州島中部，比最早的石片技術(shù)晚了約1000 年。雖然石葉技術(shù)可能起源于前一時期的長石片技術(shù)，但突然性的同時出現(xiàn)也可能意味著這一技術(shù)由朝鮮半島擴散而來。本文認(rèn)為，由于區(qū)域適應(yīng)策略的不同，源自朝鮮半島的石葉技術(shù)進入到古本州島東北部，包括本州西部的日本海沿岸地區(qū)，而不是石片技術(shù)長期盛行的西南地區(qū)。

關(guān)鍵詞：日本列島；舊石器時代晚期早段；遷徙路線；臺形石器；鋸齒刃器；石葉技術(shù)

Abstract： The beginning of the Japanese Upper Paleolithic has mainly been examined using twomajor models： the Middle Paleolithic evolutionary model within the archipelago and the continentalUpper Paleolithic diffusion/migration model. However， recent archeological data from Japanand nearby countries are challenging such simple models. This paper critically reviews previous chronology of the Japanese Paleolithic， including possible Lower and Middle Paleolithic （LP/MP），and attempts to show an alternative model of the beginning of the Japanese Upper Paleolithic. Thispaper suggests several possible specimens of LP/MP and recommends further geoarchaeologicalinvestigation to understand the reliability and cultural relationship between possible LP/MP specimensand the Early Upper Paleolithic （EUP）. The start of the Japanese EUP is presently characterized bya flake industry with trapezoids and denticulates around 39-37 kaBP cal on Paleo-Honshu Island，which has partial resemblance with contemporary assemblages in China and the Korean Peninsula，although trapezoids are endemic only to the Japanese EUP and may have derived from the ancestrallithic tradition. Blade technology appeared earliest on Central Paleo-Honshu Island， about 1000years later than the earliest flake technology. Although blade technology may have originatedfrom the elongated flake technology of the previous period， the sudden simultaneous emergenceimplies that it diffused from the Korean Peninsula. This paper proposes that blade technologyfrom the Korean Peninsula arrived on the northeastern Paleo-Honshu Island， including the JapanSea coastal region of western Honshu， rather than the southwest， where flake technology longprospered， due to differences in ecological settings and adaptation strategies between the two regions.

Keywords： Japanese Archipelago; Early Upper Paleolithic; migration route; trapezoid;denticulate; blade technology

1 Introduction

Although it is certain that modern human migration waves occurred at several timesthrough multiple routes into East Asia[1-3]， eventually reaching the Japanese Archipelago[4-7]， thedetailed formation process of the Upper Paleolithic （UP） culture has not been elucidated enoughin Japan. Studies on the beginning of the Japanese UP have mainly followed two major models：the MP evolutionary model within the Japanese Archipelago[8] and the continental Blade Industrydiffusion/migration model [9-11]. However， recent archeological data from Japan and othercountries have challenging such simple models [5-7，12].

This paper critically reviews the past chronology of the Japanese UP to examine eachpossible migration route， focusing on recent archeological and anthropological evidence in andaround the Japanese Archipelago， and attempts to show an alternative model of the beginningof the Japanese UP. In addition， this paper re-examines possible Lower Paleolithic （LP） and MPspecimens in the Japanese archipelago.

2 Paleogeography of the Pleistocene Japanese Archipelago

The UP in the Japanese Archipelago began with the late Marine Isotope Stage （MIS） 3，characterized by a colder and drier than the early MIS3， towards Last Glacial Maximum （LGM） [13-14].

Currently， the Japanese Archipelago consists of four major islands， Hokkaido， Honshu，Shikoku， and Kyushu， and more than 6，800 small islands， though recent studies of glacialeustatic changes in sea level[15-16] and sea bottom topography of channels[17-18] along the JapaneseArchipelago revealed various landmass situations during glacial period as follows （Fig.1） [19].

During the late MIS3 and MIS2， the Sakhalin， Hokkaido， and the southern Kurile Islandswere combined into a sole Paleo-Sakhalin-Hokkaido-Kurile （Paleo-SHK） Peninsula connectedto the Asian continent， whereas glacial eustasy caused Honshu， Shikoku， and Kyushu to forma single landmass called the Paleo-Honshu Island[20-23]. A chain of islands， which are presently theRyukyu Islands， was separated from each other also in the Pleistocene， extending to the southwest ofPaleo-Honshu Island. This paper refers to these islands as the Paleo-Ryukyu Islands [24-25].

In considering Pleistocene human cultural dynamics in Japan， it is important to note thatthe narrower Tsugaru Strait then the present exist during the glacial period， separating Paleo-Hokkaido and Paleo-Honshu; therefore， these two landmasses can be suggested to hold largelydifferent cultural dynamisms， at least in the UP. However， there were interrelationships oftechnology， information， culture， and human groups between the two landmasses[21-22，26-27]. Inaddition， the narrower Tsushima Strait was located between Paleo-Honshu and the KoreanPeninsula， a part of the eastern coast of the Asian Continent in the Pleistocene[18]. This gapbetween the Continent and the Paleo-Honshu caused differences in the formation process ofhuman cultures between neighboring regions as well. On the other hand， the Paleo-RyukyuIslands have often been argued to be closely related to Taiwan， southern China， and SoutheastAsia during the Pleistocene based on biogeographic evidence [28-29].

3 Possible Lower and Middle Paleolithic specimens

Several Paleolithic sites across the Japanese archipelago may have a possibility to be older than 39.0 kaBP cal， predating the UP[8，24，30].

Yoshikawa et al[31] discussed the Quaternary biostratigraphy of proboscidean fossils andclaimed proboscidean migration occurred in two stages： Stegodon orientalis around 0.63Ma（MIS16） and Paleoloxodon naumannni around 0.43Ma （MIS12）. In addition， the Paleo-HonshuIsland was last connected to the Korean Peninsula for a short period at the beginning of the LatePleistocene around 120-130 ka. Potential LP and MP sites may have been formed by humanswho migrated during these times[32].

Tab.1 shows a tentative chronology of the Japanese LP and MP， summarizing chronologicalposition and estimated age on the basis of geochronology and radiometric data. It should be notedthat the boundary between the LP and MP in Japan was tentatively set at the beginning of theLate Pleistocene. At present， the Kaseizawa site in Aichi Prefecture is the only site that was datedto the Japanese LP[33]. However， these specimens lacked geochronological context， as they werecollected after the site was destroyed by construction work. The Kaseizawa assemblage consistedof pointed pebble tools made on a flat pebble （Fig.2）， various flake tools， and a discoidal core，including a combination of large and small tools.

Other sites were assigned to the MP， as they were estimated to be younger than 120 kaBP basedon thermo-luminescence dating and tephrochronology. Based on widespread tephras， such as K-Tz （95 kaBP）， Aso-4 （90-85 kaBP）， and AT （30-29 kaBP） from south Kyushu and DKP （55 kaBP） fromthe Chugoku Mountains， which provide the chronological basis of the MP， Sato[19，30，34-35] dividedthe Japanese MP into three stages： early MP dated between 120-60 kaBP， late MP dated between60-50 kaBP， and transition from the MP to UP dated between 50-40 kaBP.

The Kanedori[36-37] and Kashiyamatate sites[38] located in Iwate Prefecture were found in amore reliable context. Both had assemblages of two different stages recovered from stratigraphicsequences （Fig.2）. Among the four artifact bearing layers of Kanedori site， “Cultural Layer 4”included Aso-4 tephra， including semi-bifacially retouched ax-like tools， pebble tools， and flakes，which have possibility to belong to the early MP[19，30]. “Cultural layer 3”， overlaying “CulturalLayer 4”， yields axes or pointed chopper， discoids， and scrapers， which may be assigned to the lateMP typologically， later than “Cultural Layer 4”. “Cultural layer 4a” of Kashiyamatate site mayhave included Aso-4， implying the same tephrochronological position as “Cultural Layer 4” ofthe Kanedori site， which consisted of flakes. “Cultural Layer 2c”， lower of the Kashiyamatate sitewas dated to the late MP， and the assemblage composed of convergent flakes and side scrapers[8，38].

Whether these MP assemblages directly or indirectly relate to subsequent UP stone tooltradition remains under debate.

4 Routes and timing of modern human migration to the JapanesAe rchipelago

4.1 Northern route

About 10，000 UP sites， including surface collections[39]， were discovered across theJapanese Archipelago， strongly suggesting that the timing of major population migration intothe archipelago was the UP. The emergence of Early UP （EUP） sites throughout the JapaneseArchipelago reflects the dispersal of modern humans into these islands[4]. This paper reviewspossible main migration routes and their evidence[40-41] （Fig.3）.

In southern Paleo-SHK （Hokkaido）， the oldest known Paleolithic occupations with securegeochronological ages were found at the Wakabano-mori site （Fig.1： 17）， Kyu-shirataki 3 site（Fig.1： 18）， and Shukubai-kaso site （Fig.1： 19）[27，42-44]. However， their radiocarbon ages are as oldas 30 kaBP cal. Izuho et al[45] suggested that the Wakabano-mori site may be older than the others.These sites are characterized by a small flake industry with trapezoids typologically comparableto the EUP in northeastern Paleo-Honshu[6，27]. Recent investigations of surface collections andexcavations at the Akita 10 site （Fig.1： 20）[46] have also yielded trapezoids and edge-ground axespertaining to the Paleo-Honshu EUP technological tradition[47]. These data imply that the terminalEUP population from northern Paleo-Honshu migrated into Hokkaido before ca.30 kaBP cal;however， there are currently no reliable radiocarbon ages available older than that. It is reasonableat present that the oldest occupation in Hokkaido （southern Paleo-SHK） may have startedsomewhat earlier than 30 kaBP cal[22，43].

The small flake industry was followed by the backed point industry， and later by themicroblade industry. Backed points were made on a blade and highly standardized withintensive retouching. The microblade industry first appeared throughout southern Paleo-SHKaround 26-25 kaBP cal and is characterized by the existence of various microblade core typesand reduction sequences[44，48-49].

4.2 Western route

At present， the number of EUP sites is estimated to be around 500[4]， with the majorityon the Paleo-Honshu Island. The origin of the EUP sites on the Paleo-Honshu Island has beenestimated at ca. 38 kaBP cal on the calibration curve of IntCal09 and 13[2，4，13，50]. In addition，Morisaki et al[7] gathered detailed stratigraphic and provenience data， lithic raw materials， tooltypes， reduction technologies， and radiocarbon dates from 42 EUP sites， demonstrating thatEUP flake reduction assemblages represented by trapezoids， scrapers， and edge-ground axesemerged ca. 38–37 kaBP cal using IntCal13. Moreover， Morisaki et al[12] obtained radiocarbondates of eight charcoal pieces from four charcoal concentrations in the earliest assemblage in thearchipelago， the Ishinomoto site in Kumamoto Prefecture （Fig.1： 57） [51]， the southwestern edgeof the Paleo-Honshu Island， which yielded trapezoids， denticulates， and pointed tools. The agesranged between 38.4-36.3 kaBP cal using IntCal13 （Tab.2）. Based on IntCal20， the age range ofthese dates was calibrated between 39-37 kaBP cal.

Another early EUP site is the Idemaruyama site （Fig.1： 43） located in Shizuoka Prefecturein the Pacific coastal region of the central Japan[52]. This site provided a similar age range tothe Ishinomoto site （Tab.2） and yielded trapezoids and a few basal retouched elongated flakes，illustrating that the earliest EUP sites are characterized by trapezoids， which are endemic to theJapanese EUP[12，26].

Re-calibration of ages of the 42 EUP sites examined by Morisaki et al[7] using IntCal20demonstrated that the oldest EUP blade reduction appeared after ca. 37 kaBP cal. The earliestexamples were recognized in the Chubu and Kanto regions on the eastern edge of centralPaleo-Honshu. These blades were detached from a flat platform on the narrow side of thickflake cores or split cobbles. Platform preparation and rejuvenation were never or rarelyperformed. Among the numerous blades produced， suitable elongated triangular blades weremodified into pointed forms by retouching their bases. There were few blade tools other thanpointed blades[7].

Following this blade production technique， another technique appeared ca. 34 kaBPcal in the Chubu and Kanto regions and was adopted in other regions ca. 32 kaBP cal， whichknocked off blades from the wide face of a prismatic or semi-cylindrical core with a preparedplatform. The produced blades were parallel-sided and showed a flat distal end with feathertermination. Some of them were modified into backed points and others into blade tools， such assidescrapers[7].

In the Japanese Archipelago， the transition from the EUP to the later half of UP is thoughtto have occurred around a huge volcanic eruption of the Aira Cardera in the southwestern edgeof the Paleo-Honshu Island ca. 30-29 kaBP cal[53]， corresponding to the transition from MIS3 toMIS2. This tephra， called the Aira-Tn （AT） tephra， covered most parts of the Archipelago， exceptfor some parts of the Ryukyu Islands and Hokkaido， providing a key to tephrochronologicalidentification of Japanese EUP sites[54].

4.3 Southern route

The Ryukyu Islands were divided into three geographic regions： north， central， and south.Kaifu and Fujita et al[28] reviewed archeological and anthropological data across the Paleo-Ryukyu Islands， showing a general perspective on Pleistocene seafaring. Fujita et al[29] recentlyillustrated the migration timing of Pleistocene islanders and their subsistence based on ampleorganic materials.

Among North Ryukyu Islands， Tanegashima Island was connected to the southern edge ofPaleo-Honshu by a land bridge due to marine regression. Two EUP sites are representative onthis island， Tachikiri and Yokomine C （Fig.1： 60， 61）， where small flake tools such as trapezoids，edge-ground axes possibly used for wood processing， and pebble tools for nut grinding werecollected[55]. These tools have a clear resemblance to the lithic tool tradition of the southwesternpart of the Paleo-Honshu Island[41，56]， indicating that these were left by people from the presentKyushu Island.

Limestone is widely formed across the Ryukyu Islands， particularly in the southwesternarea. Unlike other parts of the Japanese archipelago， karstic caves in the Ryukyu Islands are wellsuited for preserving bone remains and contain many Pleistocene fossil sites[29]， although thereare no indisputable lithic tools of this period in islands south of Okinawa.

On Central Ryukyu Islands， several archeological and anthropological sites have beenexcavated[29]. Several human skeletal remains from the Pleistocene were recovered fromMinatogawa fissure （Fig.1： 62）， located on the southern edge of Okinawa Island. Although theremains lacked bone collagen， their ages were estimated to be between 29-22 kaBP cal based onradiocarbon dates of two charcoal samples from the fissure. There were no associated artifactsfound. The oldest human remains was unearthed in Yamashitacho Cave 1 （Fig.1： 63）， locatednear the southwestern part of Okinawa Island. Charcoal from a lens deposit directly above thehuman remains was dated as 32.1±1 kaBP in 14C （TK-78）[28]. Despite the wide standard error，the calibrated age ranged from 39，430-34，670 BP cal using the IntCal20 calibration curve.Surprisingly， recent excavations conducted at Sakitari Cave （Fig.1： 64） uncovered a uniqueassemblage of shellfish products， including items that the excavation team considered fishhooks，attributed to 23-20 kaBP cal[29].

Two representative anthropological sites are known on South Ryukyu Islands. Human boneswere accidentally found in Pinza-abu Cave in 1979 （Fig.1： 65）， and additional remains wereexcavated in the early 1980s. Human and other animal fossils were dated to 25.8±0.9 kaBP in 14C（TK-535） and 26.8±1.3 kaBP in 14C （TK-65）， respectively[28-29]. These dates were calibrated as ca.32-28 kaBP and 35-29 kaBP cal， respectively， using IntCal20.

A large number of human bones comprising at least 19 individuals were recently excavatedin Shiraho-Saonetabaru Cave on Ishigaki Island （Fig.1： 66）， with no accompanying stone tools[57].A total of 104 radiocarbon dates ranging from 28.5 kaBP cal to modern dates， including age gapsof approximately 13-11 and 8-5 kaBP cal， were reported. Dates from bone concentration fromcliff burials fell between 28-20 kaBP cal[57-58].

Pleistocene human culture in Central and South Ryukyu Islands may have differed fromNorth Ryukyu Island. Refering to recent genetic analysis[59] illustrating that the weak relationshipbetween Pleistocene and Holocene populations of the Ryukyu Islands， Fujita et al[29] suggestedthat the Paleolithic people who lived in Paleo-Ryukyu may have made little or no geneticcontribution to the present population of Okinawa.

4.4 Summary

Evidence indicates that each migration route opened at different times， implying that Paleo-SHK was first inhabited by people from Paleo-Honshu Island ca. 30 kaBP cal， during the terminalEUP. The northern route seems to have been first used ca. 26-25 kaBP cal by a population fromSiberia equipped with composite microblade tools that targeted mammoth fauna. The southernroute opened no later than 35-34 kaBP cal， although the migrants did not possessed lithic tooltradition clearly pertaining to that of the Palaeo-Honshu， and the population size that entered into the Palaeo-Honshu through the southern route is still open to discussion at present. The oldestand main migration route into the present Japanese Archipelago is thought to have been thewestern route， through which the earliest UP population migrated into Paleo-Honshu and resultedin various EUP lithic traditions thereafter.

5 Discussion： the beginning of EUP in Paleo-Honshu Island

5.1 The beginning of EUP

Modern humans arriving via the western route marked the beginning of the Japanese UPon Paleo-Honshu Island. This paper discusses the origin of the EUP on this island using lithictechno-typology data.

The earliest EUP assemblage， represented by trapezoids and denticulated scrapers， datesto 39-37 kaBP cal （Fig.4 bottom）[12]. These assemblages predate the appearance of the bladeindustry in Central Honshu; however， several behavioral features， including the existence ofregional tool types， such as trapezoids and edge-ground axes， possession of seafaring technology，and exploitation of offshore obsidian of Kozushima， strongly indicate that they were left bymodern Homo sapiens[2，60]. The site distribution of this stage is limited to the southwestern Paleo-Honshu Island.

The resemblance between denticulated tools from Paleo-Honshu Island and the KoreanPeninsula and China has been previously pointed out[16，61]. A recent review of archeological datafrom China[62-63] summarized the core-flake technocomplex with denticulates typified by moresimply organized flake reduction and lightly retouched tools dominated from MIS4 to earlyMIS2， whereas Initial UP （IUP） assemblages tentatively invaded northwestern China from theAltai region. In addition， Nagai[64] showed that flake tools and pebble tools were dominant incontemporary assemblages in the Korean Peninsula.

As EUP lithic tools， particularly denticulated ones， exhibit partial similarity withcontemporary assemblages in China and the Korean Peninsula， diffusion/migration from thoseregions may have been responsible for the formation of the Japanese Earliest UP， while theexistence of trapezoids， an endemic tool type in the archipelago， likely derived from the ancestrallithic tradition.

5.2 Appearance of Blade Industry

Several possible hypotheses on the origin of Japanese EUP blade technology should bediscussed based on solid evidence. At present， EUP blade technology in the Japanese Archipelagoappeared around 37-36 kaBP cal on Central Paleo-Honshu Island， such as the Happusan II site（Fig.1： 38）[7，65]， although there are no contemporary examples to the west. On the other hand，elongated flake technology in the preceding assemblage is also a possible candidate for the archetype of EUP blade technology[12]. However， as the appearance of blade production seemsto have been sudden， it may have derived from the nearest continent by subsequent waves ofdiffusion/migration[7]. Nevertheless， it is unlikely that IUP in the Altai and eastward was directlyresponsible for the origin of the Japanese EUP[7，57]， as the total reduction sequence includingtoolkit variation of IUP in Altai[66]， Mongol[66]， Zabaikal[67]， and northwestern China[1，63] were not observed at the Happusan II site and other contemporary sites， despite that the appearance ofsome similar typo-technological features. Laminar technology of the Altai IUP， for instance， wasrepresented by the production of large and robust blade blanks and retouched blade tools， whereasthick technical side blades were turned into burin-cores to detach small laminar blanks[66]. Largeblades were used as blanks for several tool types. This complete set of the Altai IUP sequenceand diagnostic tool types seems quite rare or absent in the Japanese EUP， indicating that therewas no direct lineage but rather an indirect relationship， if any.

Sato[5] and Morisaki et al[7] presented the hypothesis that multiple diffusion/migrationconcerning blade technology in the Korean Peninsula caused the development of Japanese bladetechnology. While the influence of the IUP around Altai on the Japanese EUP is not completelydenied， this paper proposes that， although the IUP itself was not found on the Korean Peninsula，contemporary blade traditions on the peninsula could have been influenced by the IUP.

Among possible earliest EUP sites， tanged （stemmed） points on blades are known as adiagnostic tool type in the Korean Peninsula[68-69]. The oldest tanged point was recovered atthe Yongho-dong site （38.500±1，000 BP 14C）[70-71]， although it is uncertain whether the dateis associated with the tanged point. Other dated tanged points made on elongated flakes werefrom cultural layer 2 of the Hwadae-ri site at ca. 38–33 kaBP cal （31，200±900 BP in 14C）[71-72]，which also yielded typical UP tools， including endscrapers， sidescrapers， notches， and blades.An assemblage from the Songam-ri site[73] comprised tanged points on blades or elongatedflakes， sidescrapers， and blades， dated to ca. 38–35 kaBP cal （33，190±160 BP， 32，300±160 BPin 14C）. Recent excavations at the Suyanggae site loc.6 （Fig.4： top） uncovered many oldesttanged points on blades dated to 44-39 kaBP cal[74]， demonstrating that base-retouched bladepoints of Happusan II， the earliest EUP blade example in Central Japan， were influenced by theKorean EUP blade tradition （Fig.4： middle）[5].

Nevertheless， it has long been enigmatic that the appearance of blade technology precededCentral Paleo-Honshu Island and was delayed in western Paleo-Honshu Island， inviting thequestion of how blade technology diffused.

To answer this question， ecological and geographical context of the Japanese UP must beconsidered. Morisaki[12，75-76] revealed that regional differences in lithic technology existed betweenthe northeast and southwest regions from the UP to the Holocene （Fig.5）. Moving north to south，LUP assemblages at Paleo-SHK were characterized by microblade technology with compositetools， blades， and bifaces， while northeastern regions of Paleo-Honshu Island had basal retouchedblade points， flakes， and bifaces and southern regions had assemblages consisting of flakeor blade backed points， blades， and bifaces. The southern part of the Ryukyu Islands lackedindisputable lithic tools. These regional lithic technology differences broadly correspond to thebiogeographic division of the archipelago， indicating regional adaptation to different resources. Asimilar pattern was observed for early pottery adoption process[12].

It is important to note that the boundary of Paleo-Honshu was not a north-south line splitting the region into east and west but rather an east-west latitudinal line extending along themountain range， which rises from the east to the west on Honshu （Fig.5）. The two sub-regionsof the Paleo-Honshu Island were both contiguous to the Korean Peninsula during the low standstage of the sea level， making it possible to directly approach from the Korean Peninsula both tothe northeast and southwest of the Paleo-Honshu Island by crossing the narrow strait penetratingthe widely exposed continental shelf[18]. It should also be noted that blade technology dominatedduring the UP in regions along the Japan Sea coast， north of the border， whereas flake technologywas prominent among regions south of the borderline， such as Setouchi， Shikoku， and Kyushu.

These data suggest diffusion/migration and the appearance of the earliest blade technologyfrom the Korean Peninsula， which was only adopted in the northeast of Paleo-Honshu Island，while residents of the southwest placed importance on simple retouched flake technologythroughout the UP （Fig.5）. Modern humans equipped with blade technology from the peninsulamay have traveled to regions with similar ecological conditions to their origin rather than clearlydifferent conditions.

6 Conclusion

This paper reviewed archeological data from the beginning of the Paleolithic in the JapaneseArchipelago， identifying several possible LP/MP sites. Further geoarchaeological investigation isrequired in order to understand the reliability and cultural relationship between possible LP/MPsites and the EUP.

The start of the Japanese EUP was characterized by a flake industry with trapezoids anddenticulates around 39-37 kaBP cal on Paleo-Honshu Island， demonstrating partial resemblancewith contemporary assemblages in China and the Korean Peninsula. However， trapezoids wereendemic to the Japanese EUP and may have derived from the ancestral lithic tradition. Theseassemblages are thought to pertain to modern Homo sapiens. The first modern human migration intothe Paleo-Ryukyu Islands likely occurred later than migration into Paleo-Honshu Island， no laterthan 35-34 kaBP cal. People on the Central and South Paleo-Ryukyu Islands lacked stone tools，using shell tools instead， indicating different cultural traditions from Paleo-Honshu Island. SouthernPaleo-SHK was populated from Paleo-Honshu Island ca. 30 kaBP cal， during the terminal EUP.

Blade technology first appeared on Central Paleo-Honshu Island about 1，000 years laterthan the earliest flake technology. Although blade technology may have originated from theelongated flake technology of the previous period， the sudden simultaneous emergence impliesdiffusion from the nearby Korean Peninsula.

This paper investigated lithic technology throughout the Japanese UP and paleo-geographicsituations and hypothesized that blade technology from the Korean Peninsula reached onlynortheastern Paleo-Honshu Island. The southwest， where flake technology prospered， was leftunaffected due to differences in adaptation strategies between the two regions. This requiresfurther investigation; however， it is crucial that historical events of the Japanese Archipelagobe explained considering the uniqueness and biogeography of the ecosystem. Our future goal isto establish a secure geochronology of archeological sites and evaluate the regional adaptationhistory of each area.

Acknowledgement：We are grateful for Drs. Yosuke Kaifu， Dai Kunikita for their help inpreparing this paper.

References

[1] Li F， Vanwezer N， Boivin N， Gao X， et al. Heading north： Late Pleistocene environments and human dispersals in central and eastern Asia. PLoS ONE， 2019， 14（5）： e0216433

[2] Kaifu Y， Izuho M， Goebel T. Modern human dispersal and behavior in Paleolithic Asia： summary and discussion. In： Kaifu Y， Izuho M， Goebel T，Sato H （Eds.）. Emergence and Diversity of Modern Human Behavior in Palaeolithic Asia. Texas： Texas Aamp;M University Press， 2015： 535-566

[3] Zwyns N， Paine CH， Tsedendorj B， et al. The Northern Route for Human dispersal in Central and Northeast Asia： New evidence from the site of Tolbor-16， Mongolia. Scienctific Reports， 2019， 9（1）： 11759

[4] Izuho M， Kaifu Y. The appearance and characteristics of the early Upper Palaeolithic in the Japanese Archipelago. In： Kaifu Y， Izuho M， Goebel T，Sato H （Eds.）. Emergence and Diversity of Modern Human Behavior in Palaeolithic Asia. Texas： Texas Aamp;M University Press， 2015： 289-313

[5] Sato H. Recent advances of the Initial Upper Palaeolithic study in Asia： Estimating the lithic assemblages of third and fourth cultural layers in the Locality 6 of the Suyanggae Site. Journal of the Korean Palaeolithic Society， 2017， 35： 5-20 （in Japanese with English abstract）

[6] Sato H. Formation process of territoriality and prehistoric societies in Upper Palaeolithic Kyushu and Hokkaido. In： Yoshidome Hidetoshi shi Tsuito Ronshu Kankokai （Ed.）. Isekigaku no Chihei [Perspectives on Archaeological Site Research]. Fukuoka：Yoshidome Hidetoshi shi Tsuito Ronshu Kankokai， 2020： 1-8 （in Japanese）

[7] Morisaki K， Sano K， Izuho M. Early Upper Paleolithic blade technology in the Japanese Archipelago. Archaeological Research in Asia， 2019， 17： 79-97

[8] Anzai M， Sato H. Transition from Middle to Upper Palaeolithic in Japan. In： Proceedings of the International Symposium ‘Chronostratigraphy of Paleolithic of North， Central， East Asia and America （Paleoecological Aspect）. Novosibirsk： Soviet Scientific Academy Siberian Branch， 1990： 97-105

[9] Matsufuji K. Study of the Origin of the Upper Palaeolithic Culture in the Japanese Archipelago. Osaka： Kyowa Press Inc， 2004： 1-143（in Japanese with English abstract）

[10] Tamura T. Neighbors， not auslanders： some thoughts on the appearance of laminar technologies in the Japanese Archipelago.Quarterly of Archaeological Studies， 2015， 61（4）： 24-44 （in Japanese with English abstract）

[11] Kunitake S， Sutou T， Tsutsumi T. The assemblage and the origin of the oldest blade industryin the Japanese Archipelago. Kyusekki Kenkyu [Palaeolithic Research]， 2021， 17： 125-146 （in Japanese with English abstract）

[12] Morisaki K， Kunikita D， Ikeda T， et al. Ishinomoto revisited： A chronological study on the beginning of the Upper Palaeolithic Period in the Japanese Archipelago. Kyusekki Kenkyu [Palaeolithic Research]， 2020， 16： 43-58 （in Japanese with English abstract）

[13] Kudo Y. Environment and culture history of the Upper Palaeolithic and the Jomon Period： high-precision radiocarbon dating and archaeology. Tokyo： Shinsensha， 2012， 1-373 （in Japanese）

[14] Takahara H， Hayashi R. Paleovegetation during Marine Isotope Stage 3 in East Asia. In： Kaifu Y， Izuho M， Goebel T， Sato H （Eds.）.Emergence and Diversity of Modern Human Behavior in Palaeolithic Asia. Texas： Texas Aamp;M University Press， 2015， 314-324

[15] Lambeck K， Roubya H， Anthony P， et al. Sea level and global ice volumes from the Last Glacial Maximum to the Holocene. PNAS，2014， 111-43： 15296-15303

[16] Sato H. Chinese， Korean， and Japanese Palaeolithic. In： Executive Committee of Iwajuku Forum （Ed.）. Proceedings of the 7th Iwajuku Forum. Gunma： Kasakake-cho Board of Education， 1999： 37-43 （in Japanese）

[17] Matsui H， Tada R， Oba T. Low-salinity isolation event in the Japan Sea in response to eustatic sea-level drop during LGM： reconstruction based on salinity-balance model. Daiyonki Kenkyu [The Quaternary Research]， 1998， 37： 221-233 （in Japanese with English abstract）

[18] Yoo DG， Lee GS， Kim GY， et al. Seismic stratigraphy and depositional history of late Quaternary deposits in a tide-dominated setting： An example from the eastern Yellow Sea. Marine and Petroleum Geology， 2016， 73： 212-227

[19] Sato H. A perspective on the Middle Paleolithic study of the East Asia. In： Derevianko AP， Shunkov MV （Eds.）. Proceedings of the International Symposium “Early Human Habitation of Central， North and East Asia： Archaeological and Paleoecological Aspects”.Novosibirsk： Institute of Archaeology and Ethnography Press， 2005： 161-171

[20] Iwase A， Hashizume J， Izuho M， et al. The timing of megafaunal extinction in the late Late Pleistocene on the Japanese Archipelago. Quaternary International， 2012， 255： 114-124

[21] Sato H， Yamada S， Izuho M. Terrestrial resources and hunting activity in Upper Palaeolithic in Japanese Islands. In： Sato H， Iinuma K （Eds.）. No to Hara no Kankyoshi [Environmental History of Grassland in the Japanese Archipelago]. Tokyo： Bun-ichi Sogo Shuppan， 2011： 51-72 （in Japanese）

[22] Morisaki K， Izuho M， Terry K， et al. Lithics and climate： technological responses to landscape change in Upper Palaeolithic northern Japan. Antiquity， 2015， 89（345）： 554-572

[23] Takahashi K， Izuho M. Formative history of terrestrial fauna of the Japanese Islands during the Plio-Pleistocene. In： Ono A， Izuho M （Eds.）. Environmental Changes and Human Occupation in East Asia during OIS3 and OIS2 （BAR International Series 2352）.Oxford： British Archaeological Reports Ltd， 2012： 73-86

[24] Sato H. Trap-pit hunting in Late Pleistocene Japan. In： Kaifu Y， Izuho M， Goebel T， Sato H （Eds.）. Emergence and Diversity of Modern Human Behavior in Palaeolithic Asia. Texas： Texas Aamp;M University Press， 2015： 389-405

[25] Yamazaki S. Palaeolithic humans and coastal subsistence： a view from the margin. Palaeolithic Kyushu， 2020， 24： 223-246 （in Japanese with English abstract）

[26] Sato H. The framework of the Upper Palaeolithic cultures in the Japanese Archipelago， and Hokkaido and Kyushu Island. Kyushu Kyusekki [Palaeolithic Kyushu]， 2000， 4： 71-82 （in Japanese）

[27] Sato H. Early Upper Paleolithic Industries in Hokkaido， Japan. In： Деревянко АП （Ed.）. Проблемы Археологии и Палеоэкологии Северной и Центральной Азии. Novosibirsk： Institute of Archaeology and Ethnography Press， 2003： 243-246

[28] Kaifu Y， Fujita M， Yoneda M， et al. Pleistocene seafaring and colonization on the Ryukyu Islands， Southwestern Japan. In： Kaifu Y， Izuho M， Goebel T， Sato H （Eds.）. Emergence and Diversity of Modern Human Behavior in Palaeolithic Asia. Texas： Texas Aamp;M University Press， 2015： 345-361

[29] Fujita M， Yamasaki S， Sawaura R. The migration， culture and lifestyle of the Paleolithic Ryukyu Islanders. In： Ono R， Pawlik A，（Eds.）. Pleistocene Archaeology- Migration， Technology， and Adaptation. London： IntechOpen （open access eBook）， 2020： 1-15

[30] Sato H. Recent research on the Early and Middle Palaeolithic in Japan： an overview. Bulletin of the Society for East Asian Archaeology， 2016， 3： 29-35

[31] Yoshikawa S， Kawamura Y， Taruno H. Land bridge formation and proboscidean immigration into the Japanese Islands during the Quaternary. Journal of Geosciences， Osaka City University， 2007， 50（1）： 1-6

[32] Sato H. Human behavior responses to environmental and resource fluctuation in the Pleistocene Japanese Archipelago. In： Tanaka Yoshiyuki Sensei Tsuito Ronshu Kankokai （Ed.）. Archaeology： Is it Science？ Fukuoka： Chugoku-Shoten， 2016： 199-214 （in Japanese）

[33] Komura H. Aichi ken kaseizawa ky?sekki jidai iseki [The Kaseizawa Palaeolithic Site]. Tokyo： Genbunsha， 1968， 1-61 （in Japanese）

[34] Sato H. Emergence of east Asian Upper Palaeolithic. Ibou， 2008， 26： 2-15 （in Japanese）

[35] Sato H. Re-examination of Middle/Upper Palaeolithic transition in the Japanese Archipelago. Al-RāFIDāN， 2017， 38： 55-60 （in Japanese）

[36] Kikuchi K （Ed.）. Kanedori iseki [Kanedori Site]. Iwate： Miyamori Village Board of Education， 1986： 1-61 （in Japanese）

[37] Kuroda A （Ed.）. Kanedori iseki： dai 2·3 ji hakkutsu ch?sa h?kokusho [The 2nd and 3rd Excavation Report of Kanedori Site]. Iwate：Miyamori Village Board of Education， 2005： 1-132 （in Japanese）

[38] Kikuchi K （Ed.）. Kashiyamatate ato hakkutsu ch?sa h?kokusho [Excavation Report of Kashiyamatate site]. Iwate： Iwate Prefectural Archaeology Center， 1996： 1-295 （in Japanese）

[39] Japanese Palaeolithic Research Association. Paleolithic Sites in the Japanese Islands： A Database. Tokyo： Japanese Palaeolithic Research Association， 2010： 1-377 （in Japanese）

[40] Nakazawa Y. On the Pleistocene population history in the Japanese Archipelago. Current Anthropology， 2017， 58： S539-S552

[41] Sato H. Kyusekki Jidai： Nihon Bunka no Hajimari [Palaeolithic Period： Emergence of Japanese Culture]. Tokyo： Keibun-Sha， 2019： 1-127 （in Japanese）

[42] Naoe Y， Nagasaki J. Lithic raw material consumption strategy of the Early Upper Paleolithic in Hokkaido. Hokkaido Kyusekki Bunka Kenkyu [Paleolithic Culture Research in Hokkaido]2005， 10： 45-58 （in Japanese）

[43] Izuho M， Akai F. Geochronology of Palaeolithic sites in Hokkaido， Japan. Kyusekki Kenkyu [Palaeolithic Research]2005， 1： 39-55（in Japanese with English abstract）

[44] Terasaki Y. The Chronology of Paleolithic in Hokkaido. In： Anzai M， Sato H （Eds.）. Kyusekki Jidai no Chiiki Hennenteki Kenkyu[Regional Chronological Studies of the Paleolithic Period in the Japanese Archipelago]. Tokyo： Doseisha， 2006： 275-314 （in Japanese）

[45] Izuho M， Kunikita D， Nakazawa Y， et al. New AMS dates from the Shukubai-Kaso site （Loc. Sankakuyama）， Hokkaido （Japan）：Refining the chronology of small flake-based assemblages during the Early Upper Palaeolithic in the Palaeo-Sakhalin-Hokkaido-Kurile Peninsula. PaleoAmerica， 2018， 4（2）： 134-150

[46] Izuho M， Morisaki K， Oda N， et al. Surface collections from Akita 10 site. In： Sato H （Ed.）. Research on the Relationships between Human and Environmental Fluctuation in the Northern Circum Japan Sea Area （NCJSA） in Late Pleistocene. Hokkaido： University of Tokyo， 2011， 98-121 （in Japanese）

[47] Oda N， Morisaki K. Recent problems surrounding the Early Upper Palaeolithic site of Akita 10， Hokkaido， Japan. Kyusekki Kenkyu [Paleolithic Research]， 2016， 12： 217-225 （in Japanese with English abstract）

[48] Nakazawa Y， Izuho M， Takakura J， et al. Toward an Understanding of Technological Variability in Microblade Assemblages in Hokkaido， Japan. Asian Perspectives， 2005， 44： 276-292

[49] Sato H， Tsutsumi T. The Japanese microblade industries： technology， raw material procurement and adaptation. In： Kuzmin Y， Keats S， Shen C（Eds.）. Origin and Spread of Microblade Technology in Northern Asia and North America. Burnaby， BC： SFU Archaeology Press， 2007： 53-78

[50] Kudo Y， Kumon F. Palaeolithic cultures of MIS3 to MIS1 in Relation to Climate Changes in the Central Japanese Islands.Quaternary International， 2012， 248： 22-31

[51] Ikeda T （Ed.）. Isinomoto isekigun II [Ishinomoto Sites II]. Kumamoto： Kumamoto Prefecture Board of Education， 1999： 1-226 （in Japanese）

[52] Harada Y （Ed.）. Idemaruyama iseki hakkutsu chosa hokokusho [Excavation Report of Idemaruyama Site]. Shizuoka： Numazu City Board of Education， 2011： 1-120 （in Japanese）

[53] Machida H， Arai F. Atlas of Tephra in and around Japan （Revised Edition）. Tokyo： The University of Tokyo Press， 2003： 1-360 （in Japanese）

[54] Oda S， Keally CT. Japanese Paleolithic Cultural Chronology; Occasional Papers 2. Tokyo： Archaeological Research Center，International Christian University， 1975

[55] Shibutani A. Starch residues on stone tools from the Tachikiri， Kakuriyama， Soujiyama and Okunonita Sites： Material. Kodai Bunka [Cultura Antiqua]， 2008， 60（1）： 130-140 （in Japanese）

[56] Sato H. Three Paleolithic cultures in the Japanese Archipelago. In： Saidin M， Lee YJ， Woo JY （Eds.）. Proceedings of the 23rd Suyanggae International Symposium in Malaysia “Suyanggae and Lenggong： Prehistory Adaptation” Penang： Organization Committee， 2018. 23-33

[57] Nakaza H （Ed.）. Report on Survey Excavation of the Shiraho-Saonetabaru Cave Site. Okinawa： Okinawa Prefectural Archaeology Center， 2017： 1-201 （in Japanese with English abstract）

[58] Sato H. Pleistocene burials and cemetery： new discovery at the Shiraho-saonetabaru Cave Site， Ishigaki Island in the southernmost Japan. In： Nakamura S， Adachi T， Abe M （Eds.）. Decades in Deserts： Essays on Near Western Archaeology in Honor of Sumio Fujii. Tokyo： Rokuichi-Shobo， 2019： 357-362

[59] Sato T， Nakagome S， Watanabe C， et al. Genome-wide SNP analysis reveals population structure and demographic history of the Ryukyu islanders in the southern part of the Japanese archipelago. Molecular Biology and Evolution， 2014， 31： 2929-2940

[60] Ikeya N. Maritime transport of obsidian in Japan during the Upper Paleolithic. In： Kaifu Y， Izuho M， Goebel T， Ono A （Eds.）. Emergence and Diversity of ModernHuman Behavior in Paleolithic Asia. Texas A amp; M University Press， College Station， 2015： 362–375

[61] Sato H. Chronological significance and human behavior of the third cultural layer in the Ushiromuta Site. In Tachibana M， Sato H，Yamada S （Eds.）. Research Report of the Ushiromuta Site. Miyazaki： Kawaminami-cho Board of Education， 2002： 382-395 （in Japanese）

[62] Kato S. The Paleolithic of China： Its industries and chronology. Kyusekki Kenkyu [Palaeolithic Research]， 2019， 15： 91-105 （in Japanese with English abstract）

[63] Li F， Kuhn SL， Bar-Yosef O， et al. History， Chronology and Techno?Typology of the Upper Paleolithic Sequence in the Shuidonggou Area， Northern China. Journal of World Prehistory， 2019， 32： 111-141

[64] Nagai K. Lithic industries during the past 50，000 years in Korean Peninsula： Evaluating the radiocarbon chronology of the middle to upper Paleolithic Transition. Kyusekki Kenkyu [Palaeolithic Research]， 2016， 12： 185-206 （in Japanese with English abstract）

[65] Sudo T （Ed.）. Happusan iseki gun [Happusan Sites]. Nagano： Saku City Board of Education， 1999， 1-619 （in Japanese）

[66] Zwyns N， Gladyshev S， Tabarev A， et al. Mongolia： Paleolithic. In： Smith C （Ed.）. Encyclopedia of Global Archeology Vol.8. New York： Springer-Verlag， 2014， 5025-5032

[67] Zwyns N， Lbova LV. The Initial Upper Paleolithic of Kamenka site， Zabaikal region （Siberia）： A closer look at the blade technology. Archaeological Research in Asia， 2019， 17： 24-49

[68] Lee HJ， Lee SS. Chronology and characteristics of the Upper Palaeolithic blade tool industry in Korea. Acta Anthropologica Sinica，2019， 38（3）： 373-388

[69] Lee GK. The characteristics of Upper Paleolithic industries in Korea. In： Kaifu Y， Izuho M， Goebel T， Sato H （Eds.）. Emergence and Diversity of Modern Human Behavior in Palaeolithic Asia. Texas： Texas Aamp;M University Press， 2015： 270-286

[70] Han CG. The Yongho-dong site and Palaeolithic culture. Daejeon Munhwa， 2002， 11： 91-105 （in Korean）

[71] Seong C. Evaluating radiocarbon dates and Late Palaeolithic chronology in Korea. Arctic Anthropology， 2011， 48（1）： 93-112

[72] Choi BK， Ryu HJ. Pocheon Hwadaeri Shimteo Guseokki Yujeok [The Hwadaeri Shimteo Paleolithic Site. Pocheon City， Korea].Chuncheon： Doseochulpan Sanchaek， 2005 （in Korean）

[73] Lee YJ， Woo JY. Suyanggae： why so important？ In： Derevianko AP， Drozdov NI （Eds.）. Topocal issues of the Asian Paleolithic.Novosibirsk： Institute of Archaeology and Ethnography Press， 2014： 129–135

[74] Lee YJ， Woo JY， Lee SW， et al. Report on the Excavation of Suyanggae Site （Loc.I and VI）， Danyang. Cheougju： Institute of Korean Prehistory， 2018： 1-821 （in Korean with English and Japanese abstract）

[75] Morisaki K. Structural Changes and Regional Adaptation of Palaeolithic Society. Tokyo： Rokuichi-Shobo， 2010： 1-262 （in Japanese）

[76] Morisaki K. The evolution of lithic technology and human behavior from MIS 3 to MIS 2 in the Japanese Upper Paleolithic.Quaternary International， 2012， 248： 56-69