(1) Overview
Context
Our database is part of the ARCHIPELAGO initiative that collects archaeological and historical information on land and sea resources utilised in the Japanese archipelago [1]. The data reported here cover representations of marine resources (fish, marine mammals and shellfish) exploited by humans in the Japanese archipelago in prehistoric and protohistoric times. The chronological coverage spans the major periods of Japan’s pre- and protohistory from the Jōmon (ca. 14,500 – 1000 BCE) to the Kofun (250–710 CE) periods, and includes the Epi-Jōmon (340 BCE–CE 700), Satsumon (600–1200 CE) and Okhotsk (500–1000 CE) periods in Hokkaido in the north of the archipelago. No representations were identified from the Ryukyu Islands in the south of the archipelago, and our chronology summarised in Table 1 thus excludes the archaeological periods unique to those islands. The representations are found on clay pots, swords, dōtaku (bronze bells), haniwa (clay figurines), other clay objects, wooden boards, stone fragments, beads, bones, teeth, petroglyphs, bronze mirrors, and tomb wall decorations.
Table 1
Periodisation and approximate dates used in the database. Based on Barnes [20].
| PERIOD | GENERIC PERIOD | DATE RANGE BCE/CE |
|---|---|---|
| Palaeolithic | Palaeolithic | Up to 14,520 BCE |
| Incipient Jōmon | Jōmon/Neolithic | 14,520–10,550 BCE |
| Initial Jōmon | Jōmon/Neolithic | 10,550–5050 BCE |
| Early Jōmon | Jōmon/Neolithic | 5050–3520 BCE |
| Middle Jōmon | Jōmon Neolithic | 3520–2470 BCE |
| Late Jōmon | Jōmon/Neolithic | 2470–1250 BCE |
| Final Jōmon (southwest Japan) | Jōmon/Neolithic | 1250–970 BCE |
| Final Jōmon (northeast Japan) | Jōmon/Neolithic | 1250–400 BCE |
| Initial Yayoi | Yayoi-Kofun | 1000–800 BCE |
| Early Yayoi | Yayoi-Kofun | 800–450 BCE |
| Middle Yayoi | Yayoi-Kofun | 450 BCE–50 CE |
| Late Yayoi | Yayoi-Kofun | 50–250 CE |
| Early Kofun | Yayoi-Kofun | 50–400 CE |
| Middle Kofun | Yayoi-Kofun | 400–500 CE |
| Late Kofun | Yayoi-Kofun | 500–710 CE |
| Epi-Jōmon (Hokkaidō) | Yayoi-Kofun | 340 BCE–700 CE |
| Okhotsk | Medieval Hokkaido | 500–1000 CE |
| Satsumon | Medieval Hokkaido | 600–1200 CE |
The earliest depictions of aquatic animals in Japan date to the Neolithic Jōmon period and consist primarily of clay models of gastropod shells. Representations of marine mammals and fish began to proliferate during the Bronze Age Yayoi (1000 BCE–CE 250) and Epi-Jōmon periods, later expanding into animal-shaped haniwa (clay figurines placed on the borders of kofun tombs), and carvings/paintings on the walls of decorated tombs during the Kofun period.
Previous studies on Japanese prehistoric and protohistoric depictions of fauna have focused primarily on terrestrial animals. Most detailed are Hideji Harunari’s [2, 3] studies of the representation of animals on cylindrical haniwa. Among the animals found on Yayoi-period depictions, deer are the most common on bronze ritual bells (dōtaku), while birds are more commonly represented on ceramic vessels. Both categories have been linked to myths and rituals [2]. Other significant studies on Japanese prehistoric art have focused either on decorated material culture [4, 5, 6, 7, 8, 9, 10] or on specific localities [11, 12, 13, 14].
Because representations of fish and other aquatic animals are relatively rare in prehistoric and protohistoric Japan, there are few studies which specifically analyse such depictions. First among these is the work of Tadashi Saitō, who devoted his life to iconographic studies of decorated tombs and examined the fish iconographies found on the walls of these tombs [15]. His work is a compilation of these depictions within Late Kofun funerary contexts. Marine animals are covered in a collection of essays on prehistoric art edited by Shitara in 2006 [16]. This includes chapters focusing on decorations on Yayoi-period pottery [7], and prehistoric art from the Chubu [12], San’in [13], and Hokkaido regions [14]. Nishimoto discussed some artefacts depicting marine animals, specifically those with shark representations [17]. Yet, his treatment of these representations is not an exhaustive study of depictions and iconographies or of the relationship between humans and marine resources in prehistoric Japanese art.
The database presented in this paper is the first to extensively catalogue representations of aquatic animals and fishing scenes from prehistoric Japan (Figure 1). It consists of 120 depictions, including 108 animals and 12 fishing scenes. The zoological classification of these animals follows the interpretations made in the original publications consulted.
Figure 1
Distribution of sites listed in the database.
Spatial Coverage
The database covers the whole Japanese archipelago, although as already noted no relevant examples were identified from the Ryukyu Islands.
Northern boundary: 45.5171
Southern boundary: 24.4470
Eastern boundary: 145.8129
Western boundary: 122.9352
Temporal Coverage
Ca. 14,520 BCE to 1200 CE
The chronological framework of Japanese archaeology remains a subject of ongoing debate, particularly with respect to regional variations [18, 19]. Table 1 presents the periodisation used by Barnes [20] with amendments for Hokkaido. The table provides a broad timeline for the Japanese archipelago. Where available, more detailed dates for the artefacts are listed in the database itself.
(2) Methods
Steps
The database was developed through cataloguing depictions of aquatic (freshwater and marine) creatures in prehistoric Japan. The aim was to achieve as comprehensive a collection as possible through a systematic investigation of visual materials and, where available, the text descriptions that accompany them, whether site reports or synthetic papers.
1. Collecting Representations
The first phase involved the collection of relevant imagery through bibliographic and archival research, as well as site and museum visits. This was carried out via online and library-based consultations of archaeological reports, museum catalogues and academic publications, in addition to in-person visits to museums and sites in Japan. A key resource during the initial data collection was previous publications on prehistoric Japanese art that list many of the items in the database (see discussion of previous studies above). These publications were augmented by searches of the Comprehensive Database of Archaeological Site Reports in Japan, managed by the Nara National Research Institute for Cultural Properties (https://sitereports.nabunken.go.jp/en). This database is partly searchable in English but the Japanese version is more comprehensive and was used in this research. Keywords used in the database searches included period names, object types, media, and folk zoonyms (discussed below).
The images in this database were compiled from several sources. These include photographs taken by the authors during on-site visits (conducted with the permission of the museums concerned), scans of published site reports—primarily those issued by local government authorities in Japan—and materials collected from museum displays, catalogues, and online repositories. In accordance with fair use principles for non-profit academic research, the source for every third-party image has been meticulously documented and is duly recorded in the Excel and ODS files, with high-resolution versions being made available separately.
2. Cataloguing
In the second phase, the collected representations were catalogued according to two main criteria: the animal depicted, and the chronological context of the artefact on which the image appears. The taxonomic classification of these animals followed interpretations found in our sources. The names that were used were recorded, translated and systematised. No adjustments to the nomenclature were proposed at this stage. Potential re-interpretation of animal classifications or broader cultural contextualisation thereof are postponed to future research.
3. Functional Classification
Finally, the representations were categorised according to their narrative content: those depicting solely the animal are labelled as AA (Animal Alone), whereas those incorporating fishing scenes are marked as F (Fishing).
Sampling Strategy
The study sought to compile all examples that were initially identified as related to aquatic animals. Representations were searched for on the basis of prior identifications found in archaeological studies, site reports and artefact descriptions found in museums. However, to validate these initial classifications, the dataset underwent a critical review by co-author and zooarchaeologist Dimitra Mylona. This assessment confirmed that the simple nature of many of the depictions makes precise taxonomic identification unverifiable. Consequently, instead of speculating on scientific names, the database retains the Japanese common names used in the primary sources to classify the animal that is being depicted, with a few exceptions. The database includes the following classes of animals, categorised according to their English common name followed by their Japanese common name in brackets. Scientific names used by the sources are listed under these categories.
– Sharks (J. same): hammerhead sharks (Sphyrna spp.; J. shumokuzame), shortfin mako shark (Isurus oxyrinchus; J. aozame), blue shark (Prionace glauca; J. yoshikirizame), see Figure 2.
– Fish (J. sakana): salmon/trout (Oncorhynchus keta / O. masou; J. sake/masu), bonito (Katsuwonus pelamis; J. katsuo) and mullet (likely Mugil cephalus or related species; J. bora), see Figure 3.
– Gastropod shells: Conch shell (Family Charoniidae; J. horagai), cone shell (Family Conidae; J. imogai), Abalone (Family Haliotidae; J. awabi) see Figure 4.
– Dolphins (J. iruka)
– Killer whale (J. shachi)
– Whales (J. kujira): other species of cetacean not classed as dolphins or killer whales.
Figure 2
Three shark species identified in Japanese publications on prehistoric art: (A) Hammerhead shark (Sphyrna spp.), (B) Shortfin mako shark (Isurus oxyrinchus), and (C) Blue shark (Prionace glauca). (Illustrations from Wikimedia Commons, not to the same scale)
Figure 3
Three fish species identified in Japanese publications on prehistoric art: (D) masu salmon (Oncorhynchus masou), (E) bonito (Katsuwonus pelamis), and (F) mullet (Mugil cephalus or other related species). (Illustrations from Wikimedia Commons, not to the same scale)
Figure 4
Two gastropod shell types identified in Japanese publications on prehistoric art: (G) conch shell (Family Charoniidae), (H) cone shell (Family Conidae), and (I) abalone shell (Family Haliotidae). (Photos from Wikimedia Commons, not to the same scale)
Shark teeth and marine turtles are currently not included in the database. The Jōmon people made frequent use of shark teeth, including Megalodon fossil teeth. There are also shark teeth representations in stone, clay and other animal bone/antler materials [21]. However, shark teeth representations can be highly stylised into a simple triangle and, as a result, secure identification is often impossible. Marine turtles are excluded because their highly stylised depictions render precise identification challenging, and even their mere classification as turtles is problematic [22].
Although our database primarily relies on identifications provided in earlier research, in some cases these sources omit an explanation for the identification. For instance, academic publications refer to specific animals—such as ‘whales’ in tomb paintings—yet the visual evidence is sometimes too degraded or ambiguous to confirm whether this a cetacean at all. Such cases have been omitted from our database.
Quality Control
All data have been examined and double-checked by multiple authors. Depictions mentioned in reports and publications that could not be located or verified were omitted from the corpus.
Constraints
The primary aim of the database is to document representations of aquatic animals and fishing scenes from across the Japanese archipelago, including Hokkaido. Although there is a large secondary literature on prehistoric art in Japan, the present database is the most comprehensive and systematic corpus of representations relating to a particular category of animals (both as elements of aquatic ecology and as economic resources). Most of the sites and artefacts listed here have been published in full in the Japanese literature, although less information is available for some entries taken from museum catalogues.
One of the principal challenges encountered relates to the condition and documentation of Kofun-period decorated tombs. The often-poor preservation of pigments made it difficult to obtain clear, high-resolution images of the motifs, limiting the ability to confirm or reassess species identification. In addition, the absence of scale references in site plans and illustrations complicated efforts to determine the dimensions of the iconography. Where possible, the size of the animal depicted was recorded; in cases with multiple animals or fish on the same artefact, only the artefact dimensions were available. The database indicates which dimensions were used.
The most difficult issue with the material reported here relates to species identification. In order to employ this database to analyse past resource use, it is important to identify the species of fish, shellfish or sea mammal depicted. In many cases, this is difficult. Because the style of all depictions of fauna in Japanese prehistoric figurative art is very simple [23, 24, 25], a lack of detail often impedes precise (species-level) identification. By ‘simple style,’ we refer to representations in which the subject lacks detailed features but consists, primarily, of an outline of the animal. This outline tends to lack certain distinctive anatomical characteristics of the animal. For instance, only the tail fin may be depicted, while all other fins are omitted. We are currently developing a rubric based on anatomical characters that are typically represented in order to verify, reassess or narrow down the species classifications given in the existing literature. This will be applied in future publications.
(3) Dataset Description
The dataset is available in Excel and ODS formats comprising an extensive collection of representations of aquatic animals and fishing scenes from prehistoric and protohistoric Japan. Data was structured according to the columns listed in Table 2. Where data were unavailable, ‘NA’ has been recorded or the cells were left blank.
Table 2
List of data columns included in the database.
| FIELD | DESCRIPTION | VALUE |
|---|---|---|
| Unique ID | A numerical identifier assigned to each representation for unique referencing. This identifier ensures efficient data retrieval and sorting within the dataset | E.g., ID001, ID002, ID003 |
| Depiction type | Categorises the depiction | AA = Animal Alone; F = fishing scene |
| Image | The visual representation under analysis taken from the sources consulted | |
| Zoological Interpretation | Identification of the depicted animal based on the existing literature | |
| Media description | Description of the medium used, providing more context about the material | Pottery or pottery fragment; clay figurine; rock and tomb art; stone objects or tablets; bone or antler object; bronze and iron swords; wooden board; clay bead; sea mammal tooth; haniwa clay sculpture; bronze mirror; iron helmet; bronze bell (dōtaku). |
| Representation technique | Categorises the technique used in the creation of the representation | Incised; painted; carved; relief; sculptured; cast. |
| Functional context | A brief description of the potential functional role or cultural significance of the depiction or the object based on previous studies | Ritualistic; funerary; decorative; utilitarian. |
| Dimensions | Dimensions of the medium or the object | Recorded in centimetres. Divided into three columns (length, width and height). Where necessary, further details are given in the notes column. |
| Period | Cultural-historical period associated with each record | E.g., Middle Jōmon, Early Yayoi, Late Kofun |
| Maximum time range | The oldest possible date range for the representation | |
| Minimum time range | The most recent date range for the representation | |
| Site | The name of the archaeological site where the entry was found | |
| Site name in Japanese | The name of the archaeological site in Japanese characters | |
| Site type | Classification of the site helping to contextualise the depiction within its archaeological environment | Settlement, burial site, ritual site, mounded tomb (kofun). |
| Town/City | The modern town or city where the archaeological site is located | |
| Prefecture | The administrative region in Japan where the site is located | |
| Region | The larger geographical region of Japan in which the site is situated, i.e., Hokkaido, Tohoku, Kanto, Chubu, Hokuriku, Chugoku, Kinki, Kyushu. (As shown in Fig. 1, there were no examples from Shikoku). | |
| Latitude | The latitude of the archaeological site. Latitudes follow those given in the published site reports in WGS 84 format. In cases where geographic coordinates were not published, the location was estimated from Google Maps. | |
| Longitude | The longitude of the archaeological site. Longitudes follow those given in the published site reports in WGS 84 format. In cases where geographic coordinates were not published, the location was estimated from Google Maps. | |
| Image Source | The publication or database from which the image was obtained | |
| Reference | A citation of the source from which the data has been drawn |
Data Type
Secondary data; interpretation of data.
Format Names and Versions
Excel, ODS, zip
Creation Dates
01/03/2024 – 01/09/2025
Dataset Creators
Claudia Zancan, Martijn Knapen, Ilona R. Bausch, Mark Hudson
Language
English
License
Creative Commons CCZero.
Repository Location
DOI
Publication Date
29/09/2025
(4) Reuse Potential
Japan’s island geography, which fosters a rich diversity of aquatic life, makes the representation of marine animals and fishing scenes in its prehistory a subject of potential interest, although one that has received uneven attention in the existing literature. The depictions in our database provide a previously overlooked source of data for exploring the relationships between early societies in the Japanese archipelago and aquatic life, including animals and shells, fishing practices, and their associated cultural meanings.
More broadly, this database serves as a foundation for comparative studies on the iconography of marine life across ancient cultures. By providing a standardized typology and a searchable digital framework for previously scattered imagery, the dataset overcomes the fragmentation of regional reports, allowing researchers to directly compare Japanese trends with maritime traditions in Korea, China, and the wider region.
Integrating this dataset with other sources —specifically zooarchaeological records— will enable researchers to analyse broader patterns of cultural selectivity. Rather than simply viewing art as a reflection of diet, scholars can identify discrepancies between the species consumed (faunal remains) and those selected for visual representation, isolating specific symbolic behaviours. Furthermore, it supports research into cognitive archaeology, and the specific motivations behind the depiction of particular aquatic creatures in given locations and contexts.
There is also potential for use of the data in historical linguistics. It is widely argued that the Japanese language spread to the archipelago in the Yayoi period, replacing earlier languages with the exception of Ainu in the north [26]. This language shift involved substratum interference and the adoption of new words for a range of flora and fauna, including aquatic animals, through coining new words, lexical recycling of existing words or borrowing words from Ainu into Japanese. The current database could thus be used for further study of that process.
Acknowledgements
We thank the reviewer for detailed and helpful comments on the manuscript.
Competing Interests
The authors have no competing interests to declare.