The Formation and Transformation of 'Tsuchigumo' Representations in Ancient Japan

Please refer to this article for an overview

Reconstruction from Representations of Political Otherness to Medieval Supernatural Narratives

The Formation and Transformation of Tsuchigumo Representations in Ancient Japan:
From Political Otherness to Medieval Supernatural Narratives

Abstract

This paper examines the historical character of descriptions of 'Tsuchigumo' (earth spiders) found in the Kojiki, Nihon Shoki, and various provincial Fudoki, while also considering their relationship to the supernatural Tsuchigumo imagery of the medieval period. Although Tsuchigumo are often perceived today as giant spider monsters, in ancient texts compiled in the 8th century, they are depicted as beings residing in specific lands and positioned within political relationships involving submission to or resistance against royal authority. This paper avoids treating these descriptions as direct records of a specific 'Tsuchigumo ethnic group' and instead analyzes them as an exonymic category constructed within royal historical narratives. In doing so, it compares the 'Tsuchigumo' at Oshi-saka in the Kojiki, the 'Tsuchigumo' in the Nihon Shoki's pre-accession records of Emperor Jimmu, and the regionally diverse Tsuchigumo accounts in the Fudoki, arguing that the commonality of the name does not necessarily imply ethnic or biological identity. Furthermore, regarding the medieval 'Tsuchigumo Soshi' and the tales of Tsuchigumo extermination centered on Minamoto no Yorimitsu, this paper rejects the linear understanding that ancient political others simply 'transformed' into monsters, instead positioning the process as one where ancient representations of the other were reconstructed within medieval warrior culture, folklore, concepts of the supernatural, and literary traditions. Finally, it examines the potential for connection with recent advancements in ancient genome research, pointing out that while it is impossible to pre-emptively establish a genetic category of 'Tsuchigumo DNA,' there is research potential in analyzing population dynamics in regions where Tsuchigumo legends exist through archaeogenomics.

Keywords: Tsuchigumo, Earth Spiders, Kojiki, Nihon Shoki, Fudoki, Jimmu's Eastern Expedition, Representation of the Other, Royal Authority, Supernatural, Tsuchigumo Soshi, Ancient DNA

1. Introduction

'Tsuchigumo' are often treated as giant spider monsters when discussing Japanese supernatural culture. In particular, the tales of Tsuchigumo extermination by Minamoto no Yorimitsu and his retainers developed into folklore, picture scrolls, and Noh plays from the medieval period onward, and further into early modern performing arts and paintings, establishing the Tsuchigumo as one of the representative supernatural beings. However, this later imagery cannot be retroactively applied to the ancient period. This is because the 'Tsuchigumo' appearing in extant ancient texts are, at least in terms of their description, not depicted as giant arthropods, but as beings who occupy specific lands, form groups consisting of multiple members, and become targets of submission or conquest by royal authority.

This discrepancy implies that there are two distinct problem areas in Tsuchigumo research. The first is an issue of ancient history and literature regarding what category 'Tsuchigumo' signified in the Kojiki, Nihon Shoki, and Fudoki, and the second is an issue of folklore and representational history regarding the process by which that name became associated with supernatural spider monsters from the medieval period onward. Although both are interrelated, they should not be treated as a direct and continuous change.

Of particular concern is whether the Tsuchigumo in ancient historical sources can be considered a single, real ethnic group. Conventionally, Tsuchigumo have often been explained as referring to indigenous or local forces that did not submit to the Yamato royal authority. However, the fact that they possessed the character of 'local forces not submitting to royal authority' is logically separate from the question of whether they were a single group in ethnic, linguistic, cultural, or biological terms. Since 'Tsuchigumo' accounts exist in geographically distant regions such as Yamato, Hitachi, Bungo, and Hizen, it is necessary to examine the possibility that this name was a label for classification and representation by the royal authority.

The purpose of this paper is not to identify the 'true identity of the Tsuchigumo' as a specific ethnic group. Rather, it is to clarify what vocabulary and physical representations were used to describe people positioned outside the political order in texts describing the formation of the ancient state, and how those representations were reused in the cultural environment of later ages.

2. Historical Sources and Methodology

In this paper, I use the Kojiki (compiled in 712), the Nihon Shoki (compiled in 720), and the various provincial Fudoki (compiled in the 8th century) as the primary historical sources for examining 'Tsuchigumo' representations in the ancient period. However, I do not treat these as contemporary records of the eras they describe, but as texts that show how the past was narrated at the time of each compilation.

This methodological reservation is essential, especially regarding the articles related to Emperor Jimmu's Eastern Expedition. The events depicted in the Jimmu articles of the Kojiki and Nihon Shoki cannot be used directly as political history of the time of their narration. Therefore, the subject of this paper's analysis is not the 'historicity' of Jimmu's Eastern Expedition, but rather the vocabulary used by 8th-century royal historical narratives to explain the formation of their dominion.

Furthermore, regarding medieval Tsuchigumo, I refer to previous research on the 'Tsuchigumo Soshi' and the Minamoto no Yorimitsu-lineage Tsuchigumo extermination tales to examine their lexical and representational connections to ancient sources. However, I do not use medieval works retroactively as sources for ancient history.

In addition, I refer to ancient DNA research not as a means to prove the ethnic reality of the Tsuchigumo, but as independent natural scientific data showing regional differences and temporal changes in the population structure of the ancient Japanese archipelago. It is a basic methodological principle of this paper not to equate textual categories with genetic clusters.

3. 'Tsuchigumo' in the Kojiki: The Eighty Braves of Oshi-saka

An important early example of 'Tsuchigumo' in extant Japanese literature is the article on Jimmu's Eastern Expedition in the middle volume of the Kojiki. It records that in the process of pacifying Yamato, Kamu-yamato-iware-biko-no-mikoto arrived at the great chamber of Oshi-saka, where 'Tsuchigumo, the Eighty Braves' existed (Yamaguchi & Kamino-shi 1997).

In this passage, the characters for 'earth cloud' are used instead of 'earth spider.' It is noteworthy that the Tsuchigumo are not described as giant spiders, but as a group composed of many brave men called the 'Eighty Braves.' To eliminate them, the Jimmu side uses a strategy of holding a banquet, having many servers carry swords, and killing them all at once with a song as a signal.

The structure of this story is fundamentally different from the later supernatural extermination tales by Minamoto no Yorimitsu. What is depicted here, at least within the world of the story, is not a battle between supernatural beings and humans, but a political and military confrontation between a force expanding its dominion and a group that has not been incorporated into it.

However, it cannot be ignored that the heteromorphic element of 'having tails' has already been added. This suggests that the Tsuchigumo were not merely political adversaries, but were also physically othered in the narratives of the royal authority. Nevertheless, it is impossible to infer the physical traits of the actual groups from such descriptions. What can be confirmed here is the representational fact that the difference between 'us, the normal' and 'them, the conquered' is visualized through physical deformity.

4. 'Tsuchigumo' in the Nihon Shoki: Non-submission and Physical Deformity

In the Nihon Shoki, in the records prior to the enthronement of Emperor Jimmu, the term 'Tsuchigumo' appears explicitly. Regarding the Tsuchigumo of Takaowari-mura, it is stated that 'their bodies were short and their limbs were long, resembling pygmies,' describing heteromorphic features such as short stature and long limbs. Additionally, a legend is included that links the change of the place name from Takaowari-mura to 'Katsuragi' to the imperial army's act of making nets from kudzu vines to ambush and kill the Tsuchigumo (Kojima et al. 1994).

What is important in this description is that political non-submission and physical otherness are mutually linked. In the Nihon Shoki, the Tsuchigumo are not merely excluded as a military resistance force; their externality to the order of the royal authority is also represented through physical deformity.

Therefore, it is not appropriate to infer from this description that a 'short-statured, long-limbed ethnic group' existed in ancient Yamato. The issue that should be addressed is why political others were described as 'different humans' physically as well.

Here, we can recognize the function of royal historical narrative in converting political inclusion and exclusion into physical representation. However, one must be cautious about immediately equating this phenomenon with 'ethnic discrimination' or 'racial representation' in the modern sense. Explaining 8th-century texts using post-modern classification concepts carries the risk of confusing categories from different eras.

5. Tsuchigumo in the Fudoki and Regional Diversity

The most significant problem posed to the understanding of the Tsuchigumo as a specific, single ethnic group is their geographical distribution. The name Tsuchigumo, or similar names, appears not only in Yamato but also in the Fudoki of Hitachi, Bungo, Hizen, and others. As Nagafuji (2009) examines the Tsuchigumo articles in the Hizen no Kuni Fudoki individually, each Tsuchigumo legend has a unique relationship with the origin of local place names, land governance, and chieftain traditions in each region.

To explain this geographical spread, it is necessary to distinguish between at least two models. One is a model in which a group with a common ethnic and cultural background was distributed over a wide area, and this group was collectively called Tsuchigumo. The other is a model in which groups that were actually different in each region were collectively described as 'Tsuchigumo' based on the classification principles of the royal authority or the compilers.

It is difficult to actively prove the first model from current historical materials. There are no contemporary historical sources indicating that Tsuchigumo was a name shared as a self-identification, nor has any basis been confirmed to integrate the Tsuchigumo nationwide into a single archaeological group through a common language, common burial system, or common material culture.

Therefore, at least for analytical purposes, it is methodologically safer to treat 'Tsuchigumo' as a category used to classify others in royal or regional historical narratives rather than presupposing it as an ethnic name.

However, it is also necessary to avoid reducing it to a single explanation of 'Tsuchigumo = anti-Yamato forces.' The circumstances of the establishment and the content of the legends in each Fudoki article differ, and if individual historical materials are separated from their regional contexts, the very diversity of the meaning that the name Tsuchigumo held will be lost.

6. Was Tsuchigumo an Ethnic Name? Re-examination as a Category

From the above, when discussing 'Tsuchigumo' as a historical entity, it is necessary to separate at least three levels.

The first is the textual fact that the names 'Tsuchigumo' or 'Tsuchikumo' actually exist in historical materials. The second is the possibility that behind these descriptions lay the memories of regional groups or chieftain forces that were not easily integrated into the royal authority. The third is the hypothesis that these multiple groups constituted a single, ethnically and genetically identical 'Tsuchigumo tribe'.

The first can be confirmed through historical materials. The second is a historical hypothesis worthy of sufficient consideration. However, direct evidence supporting the third is difficult to accept at this point.

By making this distinction, the question of 'where did the Tsuchigumo disappear to' can also be reconstructed. If Tsuchigumo was not a fixed ethnic name but an exonymic category indicating a specific political relationship with the royal authority, then the name 'Tsuchigumo' could also lose its social meaning once that relationship was dissolved through submission or political reorganization.

In other words, even if the people were not physically exterminated, 'Tsuchigumo' could disappear from history.

This point is also important for understanding the formation of the ancient state. Political integration does not necessarily involve the annihilation of external groups, but rather the change of the classification system itself.

7. From Ancient Political Others to Medieval Apparitions

In the medieval period, the meaning of Tsuchigumo changed significantly. Within the tales of their subjugation by Minamoto no Yorimitsu and his retainers, the Tsuchigumo were reconstructed as supernatural monsters, and through works like the 'Tsuchigumo Soshi' and later Noh plays, their image as spider-like yokai was strengthened.

Regarding this change, Honda (2017) points out that the Tsuchigumo, who were described in the Kojiki and Nihon Shoki as local powers that did not submit to the imperial court, were reconstructed as monster-slaying tales in the context of medieval Otogizoshi samurai literature. This understanding indicates the necessity of viewing the ancient Tsuchigumo and the medieval spider-yokai not as the same entity undergoing continuous change, but as the same name and memory of 'otherness' being re-signified within different cultural contexts.

Therefore, while the proposition that 'conquered indigenous people were turned into yokai' is attractive, it is too simplistic as a historical explanation on its own. This is because there is a temporal gap of several centuries between the ancient Tsuchigumo records and the medieval monster tales, during which multiple cultural elements—such as samurai society, war chronicles, Buddhist views of the uncanny, Chinese literature, and oral traditions—intervened.

What should be noted instead is the possibility that what was maintained from the ancient to the medieval period was not the 'Tsuchigumo tribe' as an entity, but rather a representational function: 'a heterogeneous existence that appears from outside the order and must be subjugated.'

From this perspective, the medieval transformation into monsters is not a direct 'evolution' of the ancient Tsuchigumo. It should be considered the result of medieval society repurposing the representation of the 'other' formed by the ancient royal authority to fit its own political and literary environment.

8. Can Ancient Genome Research Verify the 'Tsuchigumo'?

Recent ancient DNA research is providing a much higher resolution of the population history of the Japanese archipelago than ever before. Genome research on the Yayoi and Kofun periods has enabled concrete analysis of the admixture of Jomon-derived ancestral components and continental-derived ancestral components, as well as regional and temporal differences.

For example, Kim et al. (2025) analyzed the whole-nuclear genomes of Yayoi people from the Doigahama site and examined the admixture of Jomon-related ancestral components and ancestral components closely related to the Korean Peninsula. Furthermore, Kim et al. (2026) analyzed multiple Yayoi individuals from northwestern Kyushu and showed that individuals retaining almost entirely Jomon-derived ancestry and individuals showing admixture with continental ancestry existed in the same region at the same time. These results indicate that the population structure of the Yayoi period itself was complex, both regionally and between individuals.

This finding makes it difficult to establish the concept of 'Tsuchigumo DNA.'

The reasons are twofold. First, there is no independent evidence to link the Tsuchigumo in literature to specific archaeological human remains. Second, even if multiple regional powers described as 'Tsuchigumo' actually existed, there is no guarantee that they were the same genetic group.

Therefore, a research design that aims to 'extract DNA from Tsuchigumo graves to find the descendants of the Tsuchigumo' faces a methodological problem: it attempts to convert a historical category into a genetic category without mediation.

On the other hand, more limited research topics are feasible. For example, it is possible to conduct genome analysis on ancient human remains from the late Jomon through the Yayoi and Kofun periods in regions where Tsuchigumo traditions were recorded, in order to examine population influx, admixture with local groups, and changes in regional genetic structure. By comparing these results with burial systems, settlement structures, material culture, paleogeography, and literary traditions, it is possible to discuss the demographic changes that occurred during the period of royal authority formation.

What is being verified here is not 'Tsuchigumo genes,' but rather the population history of the regional society where the Tsuchigumo tradition was formed.

9. Discussion: The Political Memory of 'Deformation'

Looking across the spectrum from ancient literature to medieval arts, one characteristic phenomenon can be recognized in the Tsuchigumo representation. It is the gradual conversion of political externality into physical and uncanny externality.

However, this change cannot be drawn as a straight line. Looking at the 'Tsuchigumo with tails' in the Kojiki or the descriptions of physical deformity in the Nihon Shoki, the deformation itself already existed at the stage of ancient historical sources. On the other hand, the medieval Tsuchigumo did not simply inherit this, but acquired new meanings within the Minamoto no Yorimitsu tales and samurai culture.

Therefore, it would be more appropriate to understand this not as a biological or substantial change where 'humans became yokai,' but as the political memory surrounding human groups being translated into different representations of the 'other' in each era.

In the royal narratives of the state-formation period, this translation took the form of 'deformed people who do not submit.' In medieval samurai society, it took the form of 'monsters to be subjugated by heroes.' And since the early modern period, the visual image of the spider-yokai has been further strengthened, establishing the image of the Tsuchigumo as it is generally understood today.

If we think in this way, the central problem of Tsuchigumo research is not a binary question of 'did a Tsuchigumo ethnic group really exist?' Rather, it lies in asking how political and regional conflicts that actually existed in a certain era were remembered through language, and how that memory was transformed by later generations.

10. Conclusion

In this paper, I have temporarily detached the Tsuchigumo and Tsuchigumo articles found in the Kojiki, Nihon Shoki, and Fudoki from the yokai imagery that emerged in the medieval period, and re-examined them as representations of the 'other' surrounding the formation of the ancient state.

First, the 'Tsuchigumo' in Oshisaka in the Kojiki and the 'Tsuchigumo' in the Nihon Shoki, unlike the giant spider yokai of later ages, are depicted as groups that existed in specific regions and were discussed within the context of their political relationship with the royal authority. However, there is insufficient evidence to regard these as direct records of a specific 'Tsuchigumo ethnic group'.

Second, since Tsuchigumo traditions are distributed across multiple regions, it is necessary to consider the possibility that the name was an exonymic category used by the royal authority to classify groups in different regions, rather than a self-designation of a single ethnic group. However, one should also avoid explaining all Tsuchigumo articles uniformly as 'anti-Yamato forces,' and individual analysis of each regional tradition is required.

Third, the monstrous Tsuchigumo image that emerged in the medieval period cannot be explained as a direct transformation of ancient political adversaries into yokai. It is more appropriate to consider that the representation of 'heterogeneous beings existing outside the order,' which was formed in ancient times, was reconstructed by medieval samurai culture, tales, views on the supernatural, and literary traditions.

Fourth, identifying 'Tsuchigumo DNA' directly through ancient genome research is methodologically impossible at this stage. This is because the correspondence between names in literature and genetic groups has not been established. However, interdisciplinary research examining population dynamics in the regions where these traditions exist using ancient DNA, archaeology, and historical geography is entirely feasible and could become a new method for understanding the social background of Tsuchigumo traditions.

From the above, the historical significance of the Tsuchigumo does not lie in discovering an unknown 'indigenous people.' Rather, it lies in providing an important case study for examining how people outside the political order were classified and dehumanized during the process of state formation in ancient Japan, and how those memories were reconstructed as supernatural phenomena centuries later.

Before 'Tsuchigumo' became the name of a yokai, it may have been a term that reflected the boundary formation of ancient society, defining who was included in 'us' and who was placed outside. Future research should not treat this hypothesis as a given conclusion, but rather verify it while maintaining the mutual independence of individual historical materials, archaeological data, regional history, and ancient genomic data.

Notes

1. In this paper, 'Tsuchigumo' is not used as a term that presupposes the existence of a specific ethnic group, but is used in principle as an analytical concept referring to the representations and designations that appear in ancient literature.

2. The accounts of the Jimmu conquest in the Kojiki and Nihon Shoki are not contemporary historical records of the era they describe. This paper does not attempt to prove the historical reality of the Jimmu conquest itself, but rather analyzes the representation of the 'other' in the royal historical narratives of the 8th century.

3. The expression 'exonymic category' does not prove that the various groups called Tsuchigumo did not have a common self-perception, but rather indicates the limitations of historical materials, as such self-perception cannot be confirmed from currently available sources.

4. The term 'dehumanization' (or 'making into a different form') used in this paper does not refer to biological physical traits, but to the representational operation in which political and cultural others are described as having bodies different from the norm.

5. No historical materials have been confirmed that substantiate a genealogical relationship between the 'Tsuchigumo' of Oshisaka and the ninjas of later ages. Arguments linking the two are not historical hypotheses, but currently belong to the realm of legendary imagination.

6. Similarly, it is not possible to identify 'Tsuchigumo' as Jomon-related groups or any other specific genetic group based on current archaeological genomic data.

References

Primary Sources and Annotated Editions

Uegaki, Setsuya (ed. & trans.) (1997) Fudoki, Shinpen Nihon Koten Bungaku Zenshu 5, Shogakukan.

Kojima, Noriyuki, Naoki, Kojiro, Nishimiya, Kazutami, Kuranaka, Susumu, & Mori, Masamori (eds. & trans.) (1994) Nihon Shoki (1), Shinpen Nihon Koten Bungaku Zenshu 2, Shogakukan.

Kojima, Noriyuki, Naoki, Kojiro, Nishimiya, Kazutami, Kuranaka, Susumu, & Mori, Masamori (eds. & trans.) (1996) Nihon Shoki (2), Shinpen Nihon Koten Bungaku Zenshu 3, Shogakukan.

Kojima, Noriyuki, Naoki, Kojiro, Nishimiya, Kazutami, Kuranaka, Susumu, & Mori, Masamori (eds. & trans.) (1998) Nihon Shoki (3), Shinpen Nihon Koten Bungaku Zenshu 4, Shogakukan.

Yamaguchi, Y., & Kannoshi, T. (Eds. & Trans.) (1997). Kojiki. Shinpen Nihon Koten Bungaku Zenshu 1. Shogakukan.

Komatsu, S. (Ed.) (1993). Nihon no Emaki Zoku 26: Tsuchigumo Soshi, Tengu Soshi, Oeyama Ekotoba. Chuokoron-Shinsha.

Tsuchigumo and Medieval Setsuwa Research

Nagafuji, Y. (2009). Reading the Tsuchigumo in the 'Fudoki of Hizen Province'. Kokubungaku: Kaishaku to Kyozai no Kenkyu, 54(7), 34–41.

Sudo, M. (2002). On the Background of the Formation of 'Tsuchigumo Soshi'. Setsuwa Bungaku Kenkyu, 37, 62–80.

Bai, X. (2017). The Reception of Chinese Literature in 'Tsuchigumo Soshi'. Kyoto Daigaku Kokubungaku Ronso, 37, 1–14.
https://doi.org/10.14989/222644

Honda, Y. (2017). Tales of Yokai Extermination by the Warrior Class: The Transformation of Tsuchigumo Extermination Tales in the Medieval and Early Modern Periods. Bulletin of the National Institute of Japanese Literature, 43, 297–318.
https://doi.org/10.24619/00003259

Archaeology and Ancient Genome Research

Cooke, N. P., Mattiangeli, V., Cassidy, L. M., Okazaki, K., Stokes, C. A., Onbe, S., Hatakeyama, S., Machida, K., Kasai, K., Tomioka, N., Matsumoto, A., Ito, M., Kojima, Y., Bradley, D. G., Gakuhari, T., & Nakagome, S. (2021). Ancient genomics reveals tripartite origins of Japanese populations. Science Advances, 7, eabh2419.
https://doi.org/10.1126/sciadv.abh2419

Kim, J., Mizuno, F., Matsushita, T., et al. (2025). Genetic analysis of a Yayoi individual from the Doigahama site provides insights into the origins of immigrants to the Japanese Archipelago. Journal of Human Genetics, 70, 47–57.
https://doi.org/10.1038/s10038-024-01295-w

Kim, J., Mizuno, F., Matsushita, T., et al. (2026). Ancient genomes reveal early-stage admixture and genetic diversity in the Northwestern Kyushu Yayoi. Scientific Reports, 16, 4833.
https://doi.org/10.1038/s41598-026-34996-7

Yamamoto, K., Namba, S., Sonehara, K., et al. (2024). Genetic legacy of ancient hunter-gatherer Jomon in Japanese populations. Nature Communications, 15, 9780.
https://doi.org/10.1038/s41467-024-54052-0

Basic and Related Literature

Aoki, S., et al. (Eds.) (1997). Nihon Shinwa Jiten. Yamato Shobo.

Kannoshi, T. (Ed.) (1995). Kojiki Nihon Shoki Hikkei. Gakutosha.

Mizoguchi, K. (2013). The Archaeology of Japan: From the Earliest Rice Farming Villages to the Rise of the State. Cambridge University Press.
https://doi.org/10.1017/CBO9781139034265

Research Limitations

This paper has three main limitations. First, it is impossible to determine from the literary analysis in this paper alone to what extent the 'Tsuchigumo' in ancient texts retain memories of actual groups. Second, detailed cross-referencing with regional archaeology for the various Tsuchigumo accounts has not been conducted, and whether the term 'Tsuchigumo' functioned the same way in each region requires future individual studies. Third, although ancient genome research is progressing rapidly, the current sample size and regional distribution are biased, making it insufficient to map political and cultural categories of specific regions to genetic populations.

Therefore, the conclusion of this paper does not establish the 'substance of the Tsuchigumo', but is positioned as presenting a methodological framework for connecting literary history, archaeology, the history of representation, and archaeogenomics, while pointing out the problems of treating the Tsuchigumo as a single ethnic group or a simple origin theory for yokai.

#Tsuchigumo #ILoveJapaneseHistory #AncientHistory #AncientJapaneseHistory #Kojiki #NihonShoki #Fudoki #JinmuTosei #History #HistoryBuff #Yokai #YokaiCulture #StrangePhenomena #Folklore #JapaneseMythology #Tradition #MedievalLiterature #MinamotoNoYorimitsu #TsuchigumoSoshi #HistoryMystery #HistoricalAnalysis #HistoricalEnigmas #LostHistory #Archaeology #AncientDNA #Archaeogenomics #Kingship #Grotesque #Kyoto #Nara #NoteHistoryDepartment