The Ninja an Invented Tradition

Journal of Global Initiatives: Policy, Pedagogy, Perspective
Volume 9
Number 1 Interdisciplinary Reflections on Japan

Article 3

2014

The Ninja: An Invented Tradition?
Stephen Turnbull
Akita International University, [email protected]

Follow this and additional works at: http://digitalcommons.kennesaw.edu/jgi
Part of the Asian Studies Commons, International Relations Commons, and the Peace and
Conflict Studies Commons
Recommended Citation
Turnbull, Stephen (2014) "The Ninja: An Invented Tradition?," Journal of Global Initiatives: Policy, Pedagogy, Perspective: Vol. 9: No. 1,
Article 3.
Available at: http://digitalcommons.kennesaw.edu/jgi/vol9/iss1/3

This Article is brought to you for free and open access by DigitalCommons@Kennesaw State University. It has been accepted for inclusion in Journal of
Global Initiatives: Policy, Pedagogy, Perspective by an authorized administrator of DigitalCommons@Kennesaw State University. For more
information, please contact [email protected].

Stephen Turnbull
Journal of Global Initiatives
Vol. 9, No. 1, 2014, pp. 9-26

The Ninja: An Invented Tradition?
Stephen Turnbull
The ninja is a well known phenomenon in Japanese military culture. The popularity of the
tradition is centered on the neighboring areas of Iga and Kōka where ninja are a profitable
tourist attraction. This paper examines the historical sources on which the ninja tradition is
based to see if the pre-eminence claimed by Iga and Kōka is justified. It is shown that they
were no different from several other places in their geography or their politics and that only
one reliable account of secret warfare can be identified before 1581, the year when Iga
Province ceased to exist as an independent self-governing entity. Secret warfare was
practiced throughout Japan but this tiny area of Japan claimed a particular expertise in it
and thereby invented a tradition that is still resonant today and now has all the hallmarks
of a cult.

The ninja has become a familiar figure in popular Japanese culture as the world’s greatest
exponent of secret warfare. He infiltrates castles, gathers vital intelligence, and wields a
deadly knife in the dark. His easily recognizable image is that of a secret agent or assassin
who dresses all in black, possesses almost magical martial powers, and is capable of
extraordinary feats of daring. He sells his skills on a mercenary basis and when in action,
his unique abilities include confusing his enemies by making mystical hand gestures or by
sending sharp iron stars spinning towards them.
There is much popular support for the historical truth that is supposed to lie behind
this familiar image. It is reminiscent of the passions displayed by the members of a religious
cult, because like any cult the ninja’s loyal followers staunchly defend both his worth and
his authenticity. Yet even the most devoted fans of what might be called “the ninja cult”
will acknowledge that a certain amount of exaggeration has probably taken place.
Similarly, only the most dogged ninja skeptic would dare to argue in an equally passionate
manner that the idea is a total fabrication. The usual approach, even among scholars, is
simply to accept the original ninja myth as a genuine historical phenomenon that has for
centuries been greatly romanticized and, more recently, highly commercialized.
This modern exploitation of the ninja has proved highly profitable, eclipsing anything
derived from Japan’s other great warrior tradition of the noble samurai, to whose example
of loyalty the ninja provides a dark antithesis of secrecy and deception. The samurai have
also been subject to exaggeration and commercialization in recent years, but whereas
examples of the samurai tradition are to be found all over Japan, the modern ninja cult has
one unusual feature in that its exploitation is concentrated in a very small area. This is the
former province of Iga (now part of Mie Prefecture) and the place with which it shares a
border, an area of modern Shiga Prefecture called Kōka. The two places once had much in
common and are often linked in the historical narratives. Nowadays Iga-Ueno City has by
far the most developed ninja-related infrastructure, making it is the best place in Japan to

10

Journal of Global Initiatives

visit a ninja house and a ninja museum, to enjoy martial arts displays, and purchase a wide
range of ninja souvenirs.
The worldwide acceptance of an underlying reality for ninja has meant that is now
almost impossible to read even an academic text about Iga or Kōka without finding some
reference to its most famous sons. For example, Yūki’s account of the medieval castles in
Kōka includes references to ninja (1988, p. 121), and Yokoyama’s study of Oda
Nobunaga’s campaigns in Ise and Iga notes that in the latter area “the name of the Iga ninja
is celebrated” (1992, p. 29). More recently, Ferejohn and Rosenbluth’s War and State
Building in Medieval Japan begins with a paragraph about ninja who “existed sometime in
the mists of Japanese history” (2010, p. 1). Its authors clearly take them for granted because
they regard the ninja of Iga as,
… one manifestation of fierce and extensive resistance to encroaching armies in
the dying years of medieval Japan. Local farming communities, particularly
those in mountain valleys, armed themselves with simple weapons and guerrilla
techniques to forestall the trend towards territorial consolidation and centralized
taxation. (Ferejohn & Rosenbluth, 2010, p. 1)
This is a perfect summary of an important political trend of the times, but to link it to
the existence of ninja must not remain unchallenged, and such challenges have indeed been
mounted in the past. As Sugiyama (1974) reminded everyone, “Nowadays ninja are
regarded as the stars of Sengoku battles. However, there are very few authentic historical
records [about them]” (p. 205).
The aim of this present study is quite modest. The Iga-Kōka ninja tradition as it is
understood today began to take shape when Japan’s civil wars started to come to an end
during the early Tokugawa Period (1603-1868). There is no space here to trace its
subsequent development into a modern cult that involved the appearance of so-called ninja
manuals and ended with superheroes. I shall examine instead the historical sources that
relate to the Sengoku Period (conventionally 1467-1603, the time of Japan’s great civil
wars) to see if they provide any justification for the ninja tradition beginning in the first
place. I shall explore the issue in the light of Hobsbawm and Ranger’s (1983) classic notion
of an invented tradition, which they define as:
… a set of practices, normally governed by overtly or tacitly accepted rules and of a ritual
or symbolic nature, which seeks to inculcate certain values and norms of behaviour by
repetition, which automatically implies continuity with the past. In fact, where possible,
they normally attempt to establish continuity with a suitable historic past. (p. 1)

“The invention of tradition” is not necessarily a pejorative term. None of the examples
given in Hobsbawm and Ranger’s book have been dreamt up from nothing and they note
that, “There is probably no time and place with which historians are concerned that has not
seen the ‘invention’ of tradition in this sense” (1983, p. 4). They also identify “the use of
ancient materials to construct invented traditions of a novel type for quite novel purposes”
(1983, p. 4) that, “extends the old symbolic vocabulary beyond its established limits”
(1983, p. 7).

Stephen Turnbull

11

All these points could apply to the development of the ninja, but if the ninja are indeed
an invented martial tradition they are by no means alone, because military societies are
particularly vulnerable to the creative approach. For example, the popular image of the
pirate derives almost entirely from the American author and illustrator Howard Pyle (18531911). Pyle regarded the look of ordinary eighteenth century sailors as too dull for these
romantic figures, so he added elements drawn from the Spanish folk costume of his day,
especially headscarves knotted behind the head, large hooped earrings and wide trailing
sashes, to a somewhat mistaken notion of authentic seafaring dress (Konstam, 2011, pp.
26-27). Nor is Japan exempt from such a process. The story of the vendetta carried out by
the Forty-Seven Rōnin of Ako derives almost entirely from an unashamedly fictionalized
play that was staged shortly after the incident on which it was based (Turnbull, 2011, p. 5).
Even the modern understanding of bushidō, the virtually sacred “way of the warrior,” owes
much to Bushido: The Soul of Japan, a book first published in 1900 by a Japanese Christian
living in America (Nitobe, 1905). As bushidō is such an important concept for defining the
way of the noble samurai it is ironic indeed to consider that the cult of the underhand ninja
may predate it by almost three centuries.
I must however declare an interest. It is not uncommon to regret the excesses of one’s
youth, and to have produced a book where enthusiasm sometimes overwhelmed common
sense is part of my own history. In Ninja: The True Story of Japan’s Secret Warrior Cult
(Turnbull, 1991) I translated the historical sources that were then available and interpreted
them as I understood them at the time. However, I allowed myself to become a little overdependent on compilations of these sources, in particular the books by Sasama (1968) and
Yamaguchi (1969), neither of whom makes clear the age and reliability of the material they
include. In their books contemporary descriptions of secret warfare are juxtaposed with
accounts that were written well after the establishment of the ninja myth and therefore
liable to have been influenced by it. In this article I shall therefore re-examine the evidence
with a degree of academic rigor that may have been lacking in 1991.

The Elusive Ninja
For the purposes of this article, I propose that if the ninja have any basis in fact the
following three criteria must be satisfied:
1.
2.
3.

A unique corpus of military techniques involving secrecy existed in Japan during the
Sengoku Period.
The exercise of these techniques was confined to certain skilled individuals rather
than being spread more widely within Japanese society.
These skilled practitioners were identified in particular with Iga and Kōka, from where
they sold their services to others.

The first criterion appears easy to satisfy. Activities conducted in secret on or off the
battlefield are common to all martial societies throughout world history so it would be
surprising indeed to find that Japan, with its long military tradition, was an exception. The
difficulty lies in identifying a uniquely Japanese version of it or a unique Japanese expertise
in it. The second criterion, that secret warfare was carried out only by a highly skilled,
highly specialized and perhaps hereditary corps of elite warriors, is complicated because

12

Journal of Global Initiatives

Japan’s samurai class were precisely that: highly specialized, hereditary, and elite. The
historical ninja would therefore have to be the “super-samurai” beloved of modern ninja
movies. In that case no ordinary samurai, no matter how good he may have been at the
martial arts, could perform these techniques of castle-entry, battle-disruption, or
intelligence gathering. They were the exclusive preserve of the ninja. The third criterion,
that the ninja came from Iga and Kōka, forms the basis for most of the modern commercial
exploitation. It is also the most difficult one to accept because it defies common sense, but
it is the feature maintained most fiercely by ninja enthusiasts. During the Sengoku Period
fighting was carried out from Okinawa to the fringes of Hokkaido, yet we are required to
believe that a tiny and relatively insignificant area of central Japan produced a warrior caste
so skilled that its talents were to be widely exported on a mercenary basis.
Beginning with the first criterion of a unique secret warfare tradition in Japan, there
are enough references to the Chinese military classics in the historical accounts to
demonstrate that the daimyō were familiar with their contents, including material on spying
and undercover operations. Mōri Motonari (1497-1571) the daimyō of Aki, certainly knew
his Sonshi (Sun Zi’s Art of War) (Shimizu, 2009, p. 34). Secret warfare was therefore not
an isolated Japanese specialty, nor was it confined to the Sengoku Period. Undercover
operations are to be found throughout Japanese history, and one component of the modern
ninja cult has been to exaggerate this fact by crediting certain historical figures with being
ninja or “proto-ninja.” The legendary Prince Yamato Takeru resorts to subterfuge on at
least one occasion including dressing up as a woman, making him a ninja in some eyes
(Aston, 1972, p. 201). One particularly romantic example of secret activity from the
fourteenth century is the account in the Taiheiki of the murder of Homma Saburō by the
youth Kumawaka, who escaped by climbing up a bamboo trunk (in itself no mean feat)
and allowing it to deposit him in a place of safety (McCullough, 1959, pp. 47-49). This is
a good story that may with complete justification be described as a “ninja-like
assassination,” but the boy was not a ninja. Kusunoki Masashige (1294-1336), lauded for
his devotion to the emperor, used guerrilla tactics including booby traps and dummy
warriors (McCullough, 1959, p. 185), but to go beyond this type of statement to claim that
these activities prove that Masashige was a ninja or even the founder of a specific ryū
(school or tradition) of ninjutsu (ninja techniques) would indeed be the inventing of
tradition. Most ninja enthusiasts do not go this far; instead they accept that throughout its
history Japan possessed individuals like Masashige who were skilled in secret warfare
along with their other martial accomplishments.
Secret operations were therefore common in Japan but by no means exclusive to it,
nor were intelligence-gathering activities always successful. In 1630 the idea of a Japanese
invasion of the Philippines was revived for the last time, largely at the hands of the daimyo
Matsukura Shigemasa (1574-1630), who believed that it would provide an answer to the
Christian problem by cutting off the main source of supply of priests. He sent two
Matsukura retainers as spies to Manila disguised as merchants, ostensibly to discuss
reopening trade (Hayashi, 1954, p. 980). The Spanish were not fooled, as their records
reveal: “Although in Manilla warning of this double object had been received, this was not
made known; and they were received and regaled as ambassadors from the Tono of Arima
and Bungo. A ceremonious reception and very handsome present were given to them; but
the city was put in readiness for whatever might happen” (Blair and Robertson, 1903, pp.
245-246).

Stephen Turnbull

13

The Exclusive Ninja
The first criterion can therefore be easily dealt with. The latter two require more analysis,
and much depends on the use of words. “Men of secrecy” is a very loose translation of the
two ideographs nin and ja (or sha, as in “geisha” for example), but this is a comparatively
modern development. In the vast majority of accounts the two characters are separated by
the particle no, so that the name of the practitioner under question is read in a different
style and becomes a shinobi no mono. If the no is absent in an old account no clue is given
as to whether the original reading was shinobi mono or ninja. Outside Japan the use of the
word ninja rather than shinobi no mono has predominated, probably because ninja trips
readily off the Western tongue. Interestingly, there has been a trend in recent years among
ninja enthusiasts to prefer the term shinobi no mono or just shinobi on the grounds that the
frequency of their use in the historical accounts confirms their authenticity. The word ninja
is then reserved for the exaggerated popular development found in comic books and
movies.
The latter two criteria will be tackled in reverse order, beginning with Iga and Kōka’s
overall claim to exclusivity. It must be stated in all fairness that the Ninja Museum at IgaUeno does offer a grudging recognition that secret warfare was carried out elsewhere in
Japan, although the point is always made most firmly that the tradition was centered on Iga
and Kōka, that the very best secret operatives came from there and that any other ninjutsu
traditions were developed by emigrants from Iga-Kōka. This belief is reflected in the
wording chosen for some historical accounts, where the use of “Iga-mono” or “Kōkamono” for practitioners of secrecy rather than shinobi no mono suggests that a casual
assumption is being made by the authors that if secret warfare is being carried out then they
must have been derived from those two places in some way.
As for the mercenary element, if men from Iga and Kōka acted as mercenaries that
alone would make them unique in Japan. A mercenary is a soldier who comes from outside
the society for whom he fights, who is not part of its regular forces and who is motivated
primarily by the desire for private gain, such as the mercenary bands of Europe who would
fight for an Italian city-state one month and fight against it the next (Urban, 2006). This
model, which was very common in Medieval Europe, was totally absent from the Japanese
scene, although some Japanese warriors did serve as mercenaries overseas. Between 1593
and 1688 Japanese fighting men, most of whom were exiles and many of whom had
experience of piracy, were in the service of the kings of Siam and Cambodia, the Spanish
colonists in the Philippines and the Dutch East India Company. There is nothing
comparable within Japan itself. Rather than preferring such a casual model, the emerging
daimyō (warlords) valued loyalty and long-term commitment. The only “swords for hire”
within Japan were small groups of desperate rōnin (immortalized forever in the film Seven
Samurai), and even they would tend to become quickly integrated into a daimyō’s army.
Nevertheless, one of the earliest and most reliable accounts of a ninja-like operation
seems to include a mercenary element. The source consists of a brief mention in Tamon-In
nikki, a diary kept by Abbot Eishun of Tamon-In, a sub-temple of the Kōfukuji in Nara.
The entry for the 26th day of the 11th month of the 10th Year of Tembun (1541) reads as
follows:

14

Journal of Global Initiatives

This morning, the Iga-shū entered Kasagi castle in secret (shinobi itte) and set fire to a few
of the priests’ quarters and so on. They also burned down outbuildings in various places
within the third bailey and are even said to have seized the first bailey and the second
bailey. According to Kizawa Nagamasa’s castle commander, a man called Ukon who was
his nephew and who had only 70 or 80 men to defend the castle with, they were from Kōka
in Ōmi. (Sugiyama, 1974, p. 206)

This account has everything that one could wish for. It is a factual record made by a
dispassionate observer writing long before the ninja cult developed. The word shinobi is
present and men from Iga and Kōka take part, the latter being included within the Iga
organization. As a consequence, much has been extrapolated from these few sentences to
make the reference into the most important proof text for three crucial points: that the ninja
came from Iga and Kōka, that they had unique skills and that they exploited them on a
mercenary basis.
However, different conclusions may be reached on examining the passage carefully,
and the first consideration concerns the status of the Iga unit who carried out the raid. They
are identified using the suffix shū (military unit, troop, or company). The usual meaning of
shū at this time in Japanese history is a unit under the command of a daimyō whose kokka
(domain) consisted of a composite of separate fiefs either held directly by the daimyō and
his family or indirectly by his followers, for whom the European term “vassal” is
customarily employed. The Hōjō of Odawara, for example, controlled about 500 samurai
retainers who were identified using the term shū, a suffix added to the name of the castle
where they would be mustered in times of war. This model of local governance was one
consequence of the breakdown of central authority under the Shogun following the Ōnin
War (1467-77), but it did not apply everywhere in Japan, and Iga was one of those other
provinces. The provincial border between Iga and Kōka may have meant as little as it did
in the case of the Hōjō’s kokka, but there the resemblance ended, and their situation is
summed up by Sasama in the following words: “In Iga Province at about the time of Ōnin,
[the family of] Niki Iga-no-kami were its provincial governors, but in later generations
their authority declined and in Iga no one controlled the jizamurai (country samurai)”
(Sasama, 1968, p. 83).
In other words, the rule by the Shogun’s appointee had ended, but he had not been
replaced by a rising daimyō family. Instead the province was still divided among the
jizamurai, landowners who operated on a much smaller scale and had not entered under
anyone’s vassalage. “Iga-shū” in the Tamon-In nikki therefore refers not to a military unit
identified with a daimyō but to one drawn from the kunishū. This term can simply mean
“the people of the province,” but during the Sengoku Period it had the more specific
meaning of local warriors who possessed certain superior rights over some land and its
workers and who would both work the land and fight in armed shū as part-time samurai
(Hall, 1981, p. 24). In Elisonas’ (1991) discussion of kunishū he distinguishes them from
daimyō by using a useful analogy from medieval Europe and calling them “barons” (p.365).
When danger threatened individual shū would join forces in an ikki (a confederation or
league). Some ikki had an almost permanent status and ruled lands as a loyal confederacy,
making an area controlled by an ikki very different from one controlled by a daimyō. The
daimyō sought to become absolute masters of the land and the people within their
territories, aiming to reduce all samurai to vassalage and all farmers to tax-paying workers,

Stephen Turnbull

15

so the relationship was hierarchical and feudal. An ikki was a voluntary confederacy
organized ideally on egalitarian lines.
The mutual need for survival in the face of threats from daimyō and their rapacious
armies was usually sufficient to bind the members of an ikki together. In other places some
extra factors strengthened the bonds such as affiliation to a particular sect of Buddhism.
Pierre Souyri, who has studied the contemporary political situation in Iga, believes that the
existence of the unique ninja tradition (which he accepts uncritically) was the “extra glue”
in the local ikki that held their confederacy together. The Iga jizamurai were certainly loyal
ikki members, and their joint efforts managed to suppress the bandits that had been active
in this wild mountainous terrain. It was also in the form of an ikki that they defied Oda
Nobunaga and his son over the course of the two campaigns that ultimately led to Iga’s
downfall in 1581 (Yokoyama, 2006).
Souyri regards Iga as politically unique, but a comparison with other areas
demonstrates that Iga was no different from any other province where daimyō rule was not
the dominant force and the combined strength of the kunishū allowed its barons to preserve
the province for themselves in spite of inroads made by more powerful neighbors. Southern
Japan provides two more examples. Higo province (modern Kumamoto prefecture) was
for decades a battlefield for the three great powers of Kyushu: the Shimazu, the Ōtomo,
and the Ryūzōji, whom the local barons fought for and against in a confusing series of
voluntary and forced alliances (Araki, 1987; Ōyama, 2003). Like Iga, Southern Japan
provided mountains as places of refuge. The phrase “a domain so completely surrounded
by a broad river and complicated mountain systems must have been singularly impregnable
against invasion,” refers not to the comparison popularly made between Iga and
Switzerland (Souyri, 2010, p. 121) but the situation of the barons who held sway in the
province of Satsuma (modern Kagoshima prefecture) (Asakawa, 1929, p. 25).
The Iga-shū were therefore an independent warrior band who were fighting outside
their own province. This would appear to strengthen the identification of them as
mercenaries, but that is neither likely nor in any way necessary. The Tamon-In nikki does
not tell us who the assailant was, but other sources reveal that it was Tsutsui Junshō (15231550) who was involved in a typical local power struggle (Sugawara, 1985, p. 102). Kasagi
was quite close to Iga; by modern roads it is a mere 20 kilometers from the present-day
city of Iga-Ueno to Mount Kasagi. Adding to this the fact that in 1541 boundaries were
defined not by province but by what could be defended, such an involvement by neighbors
in a dispute is by no means uncommon, particularly when the warring neighbors are rival
daimyō. Just like the barons of Higo the Iga ikki would fight for or against a particular
daimyō when it was in their interests to do so, and the Tsutsui of Yamato were no strangers
to Iga province. In later years Tsutsui Junkei (Junshō’s son, 1549-1584) would play a role
in the final invasion of Iga and Junkei’s son Tadatsugu would become Iga’s daimyō in 1585
(Lamers, 2000, p. 204). There is therefore no need to assume that in 1541 the Iga-shū were
acting as politically disinterested mercenaries seeking only financial gain from a willing
employer. It was very much in their interests to be involved with a powerful neighbor.
Even if a purely mercenary role has to be ruled out the Tamon-In nikki still refers
specifically to particular skills in secret warfare being displayed by men from Iga, and so
does Nochi Kagami, which describes how in 1487 the Shogun Ashikaga Yoshihisa led a
campaign against Rokkaku Takayori. Yoshihisa gave battle at a village called Magari
where someone from Iga earned renown by operating in a shinobi manner. This, says the

16

Journal of Global Initiatives

author of Nochi Kagami, “is the origin of the fame of the men of Iga” (Sugiyama, 1974, p.
205). Unfortunately Nochi Kagami was compiled by the Confucian scholar Narushima
Kadō (1802-53) and published in the year of his death, so it is far too late to have avoided
being influenced by the ninja myth. There is no reason to assume that the story is made up,
so in this brief reference to someone from Iga exercising the skills of undercover warfare
we may well have the earliest inkling of the expertise which the Iga-Kōka ninja tradition
would appropriate and exploit. It also means that the Tamon-In nikki story is the only
authentic contemporary account of Iga-Kōka secret warfare that exists, a revelation that
may surprise the reader.
In the Tamon-In nikki account the contingent from Kōka are regarded as forming part
of the overall Iga-shū, and this reflects the close relationship that existed between the two
places at that time. Kōka’s political situation was also similar to Iga’s. Twenty-one
yamashiro (hilltop castle) sites have been identified in Kōka together with many more
small lookout posts, evidence of extensive but small scale military activity where the
various families were joined together in a classic ikki model that some writers have dubbed
“republicanism” (kyōwasei) (Yūki, 1988, p.117). The leading Kōka barons were so strong
that one of their number Wata Koremasa (1536–83) gave refuge there to the fugitive (and
future shogun) Ashikaga Yoshiaki after the suicide of his brother Yoshiteru in 1565 (Yūki,
1988, p. 119; Elisonas & Lamers, 2011, p. 117).
The Kasagi raid took place in 1541, and much was to change over the next half century.
Like every other Japanese province Iga and Ōmi (of which Kōka was a small part) would
experience an unstoppable transition as the movement towards re-unification took place
from about 1560 onwards. The process was begun under Oda Nobunaga (1534-1582),
completed under Toyotomi Hideyoshi (1536-1598), and consolidated under Tokugawa
Ieyasu (1542-1616). It was largely a military operation that would be enforced by land
surveys, the transfer of landowners and forcible disarmament. The final result would be a
shift from the rule by independent barons and small provincial daimyō to a national
hegemony where the only daimyō left were the appointees of the national ruler.
Throughout the time of Oda Nobunaga’s rise to power Ōmi was a battlefield, and there
are references to men from Iga and Kōka assisting local daimyō Rokkaku Jōtei in the
Shinchō-Kō ki. This is a chronicle of Oda Nobunaga’s exploits written by one of his
generals who was an eyewitness to many of the incidents he described. None, however,
involves secret warfare. In 1570 there is armed confrontation with Nobunaga in southern
Ōmi, and the Shinchō-Kō ki records that his men “killed 780 accomplished samurai from
Iga and Kōka” (Kuwata, 1965, p. 105; Elisonas & Lamers, 2011, p. 145). Three years later
Ashikaga Yoshiaki, who by now has become Shogun and has turned against Nobunaga,
has men from Iga and Kōka fighting for him at Ōtsu (Kuwata, 1965, p. 136; Elisonas &
Lamers, 2011, p. 183). Later in that same year of 1573, “skilled archers from Iga and Kōka”
are found assisting the Ikkō-ikki of nearby Nagashima. The Nagashima army was an ikki
whose members and bound together by their allegiance to the Jōdo Shin sect of Buddhism
(Kuwata, 1965, p. 151; Elisonas & Lamers, 2011, p. 201). Ōmi Province was finally
pacified when Nobunaga accepted the surrender of Rokkaku Jōtei. Nobunaga then stamped
his authority on the province by choosing Azuchi, on the shore of Lake Biwa and in the
middle of the province, as the site for his headquarters. Territories were redistributed, and
in the patchwork of landholdings that emerged the old Kōka republicanism disappeared.

Stephen Turnbull

17

Just across the border in Iga, however, the situation was somewhat different. The
strength of the Iga ikki and the surrounding mountains allowed the men of Iga to maintain
their independence for several more years, and in 1579 they beat off Nobunaga’s son Oda
Nobukatsu (1558-1630) when he was sent to punish them. It is very noticeable that they
received no help from Kōka at this time. The old cross-border alliance had passed into
history, and when Nobunaga’s army returned with an overwhelming force in 1581 and
crushed Iga the Kōka-shū were to be found fighting on his side (Kuwata, 1965, p. 332;
Elisonas & Lamers, 2011, p. 410). By New Year’s Day 1582 the Kōka-shū were so closely
associated with Nobunaga that they are mentioned in the same sentence as his elite horse
guards when “the major and minor lords from the neighboring provinces as well as the
fraternal branches and their retinues” came to visit Nobunaga in his magnificent castle of
Azuchi (Kuwata, 1965, p. 343; Elisonas & Lamers, 2011, pp. 421-422).
The belief that the Iga ninja tradition not only survived but even prospered after the
cataclysm of 1581 is one of its strangest features of the modern ninja cult (Souyri, 2010, p.
122). From this time on, the popular yet paradoxical argument goes, the ninja of Iga, who
were desperate for employment, marketed their unique ninjutsu skills and began to sell
their services as mercenaries more widely than ever before. Others emigrated to distant
provinces and set up schools of ninjutsu. There are however no authentic references to men
from Iga fighting in an independent capacity after 1581 to add to the sole reference before
that time. It is not much on which to erect so great an edifice of commercialized tradition,
and while it must of course be accepted that the invasion undoubtedly destroyed any written
records that may have existed, the important point is that there are no authentic records
after it. Oral tradition can be very powerful but I would argue for three reasons that the
condition of Iga Province after 1581 would have made it impossible for independent
mercenary warfare to flourish.
The first reason is simply that when the kunishū of Iga were wiped out the province
passed rapidly under daimyō control. Three out of the four districts of Iga were assigned to
the victorious Oda Nobukatsu and the remaining one to Nobunaga’s brother Oda Nobukane
(1548-1614) (Kuwata, 1965, p. 334). Nobukatsu survived his father’s murder in 1582 and
went on to serve Toyotomi Hideyoshi, from whom he received Owari and Ise provinces in
addition to Iga. He then made the fateful decision to fight against Hideyoshi at the battle
of Nagakute in 1584, so in 1585 Hideyoshi deprived Nobukatsu of his lands as a
punishment and transferred the Tsutsui family from Yamato to Iga. In Lamers’ (2000)
words, “they were unknown and unloved” in Iga, but that statement is only half true (p.
204). As the Tsutsui had taken part in the destruction of Iga in 1581 they were far from
being unknown to their new subjects, although they are highly likely to have been unloved.
Finally, in 1608 under Tokugawa Ieyasu’s resettlement scheme, Tōdō Takatora (15561630) was moved from Imabari to the fief of Anotsu which included both Ise and Iga
provinces. From 1581 onwards, therefore, Iga province was a daimyō domain, so that any
subsequent accounts of men from Iga in action, secretly or not, must be interpreted as
indicating their inclusion in a daimyō’s army because no other form of army could possibly
exist.
The second reason for the non-existence of ninja in post-1581 Iga is its well-recorded
state of lawlessness. Banditry, once repressed by the Iga ikki, now flourished anew as is
implied by the circumstances surrounding the well-known episode whereby Tokugawa
Ieyasu chose Iga as the route for his escape when Oda Nobunaga was murdered in 1582.

18

Journal of Global Initiatives

The usurper Akechi Mitsuhide had moved rapidly to eliminate Nobunaga’s family and
close allies, killing Nobunaga’s heir and almost trapping Ieyasu in Sakai with a small
defensive retinue. Ieyasu, however, made his way safely back to Mikawa Province by
taking a devious route through Iga on the advice of his loyal retainer Hattori Hanzō, a man
born in Mikawa from an Iga family. Ieyasu had provided refuge in Mikawa for those fleeing
from Nobunaga’s assault in 1581 so he was highly regarded in Iga, but far from confirming
the popular interpretation of an orderly province ruled by a network of ninja families this
celebrated episode shows only chaos in a place infested by bandits. (Kuwata & Utagawa,
1976, p.191).
The third consideration is that Iga must have suffered as much as any other province
from Hideyoshi’s Katanagari (Sword Hunt). This was the operation by which Hideyoshi’s
agents forcibly confiscated all weapons from anyone except a daimyō’s samurai, most of
whom were now completely separated from the land, based in castles and in a state of
vassalage to Hideyoshi’s most loyal generals. The operation began in 1585 not far from
Iga and finished nationwide in 1587 with even the inhabitants of the “holy mountain” of
Kōya-san being ordered to disarm (Fujiki, 2005, p. 59). The Sword Hunt was so thorough
that it would be incredible to think that the men of Iga, a province already devastated by
war, could have retained any weaponry, let alone had the means to use them other than in
the service of their newly imposed rulers.
The independent Iga-shū who fought at Kasagi in 1541 had therefore ceased to exist,
yet they were to live on in romanticized accounts as part of the growing ninja myth. One
contributor to it is the Taikō ki, an elaborate fictionalized version of the life of Toyotomi
Hideyoshi that was published in 1625. The relevant section concerns the invasion of Korea
in 1592 and the successful capture of the castle of Chungju, the final obstacle on the road
to Seoul. The Japanese commander Konishi Yukinaga divided his contingent of “100 Iga
shinobi no mono” into two and sent half of them to attack the rear of the castle, where they
set fire to the town outside (Yoshida, 1979, pp. 107-108). This is most unlikely to have
happened, partly for the reasons stated above but also because the Korean invasion was
very much a Kyushu-led affair. The account itself is also at variance with more reliable
descriptions of the fall of Chungu, which capitulated after its commander had made the
disastrous decision to take the garrison out of the castle and face the Japanese army in a
pitched battle with a river at his rear (Park, 1978, p. 104). Instead this passage may be taken
as evidence of how firmly the Iga ninja myth had taken root within forty years of the
destruction of the province.
After 1581, therefore, none of the reliable accounts of secret operations in Japan makes
any reference to them being carried out by warriors from the very place that posterity would
regard as their heartland. In other words, Nobunaga’s invasion of Iga did not mean the start
of the Iga ninja tradition. It meant the end of it, and in fact Iga was so thoroughly pacified
that neither samurai nor ninja from Iga province played any part in the great Sekigahara
campaign of 1600. Their daimyō Tsutsui Sadatsugu simply locked himself in Iga-Ueno
Castle and sat out the war, and his failure to provide active support to the victor Tokugawa
Ieyasu resulted in the confiscation of his lands in 1608.

Stephen Turnbull

19

The Ninja of Kōka
Within twenty years of Sekigahara the Iga-Kōka ninja myth had begun, culminating in the
highly flexible concept of the superhero ninja of modern times who can fly through the air
or disappear at will. That image is of course the product of fantasy, but the evidence
strongly suggests that during the first half of the seventeenth century a less fantastic
manipulation of the facts also took place, and because Iga had been so thoroughly crushed
much of it involved Iga’s neighbor Kōka. The first reference to Kōka in this context is
Mikawa Gofudo ki, a detailed account of the exploits of Ieyasu that was written in 1610
and edited in 1837 (Kuwata & Utagawa, 1976). The reported incident happened in 1562
when Imagawa Ujizane took members of Ieyasu’s family hostage in Sumpu Castle.
Ieyasu’s plan was to effect the rapid capture of the Imagawa’s own Kaminogō Castle and
take hostages who could then be exchanged for his own family:
The monogashira Mitsuhara Sanzaemon said, ‘As this castle is built upon a formidable
precipice we will be condemning many of our allies to suffer great losses. Fortunately there
are among the honourable hatamoto some men associated with the Kōka-shū of Ōmi
Province. Engage the men of Kōka through their kinsmen and they can enter the castle
secretly’. Tadatsugu agreed and beginning with Tomo Tarōza'emon Sukeie from Kōka
eighty men trained as shinobi were engaged and ordered to hide in various places, and on
the 15th day of the 3rd month they entered the castle in secret, and were soon setting fire
to the towers inside the castle. The attackers deliberately did not converse as they ran
around killing and the garrison thought they were traitors. (Kuwata & Utagawa 1976, p.
144).

Thanks to the men from Kōka, Ieyasu acquired the hostages he needed in the form of
the two sons of the castle keeper Udono Nagamochi, who was killed in the fighting.
Following a deal with Imagawa Ujizane the castle was handed back and the sons appointed
to its command, but the pair began an unwise policy of supporting the raids into Mikawa
Province by the Ikkō-ikki. Ieyasu again entered the field against Kaminogō and once more
the Kōka-shū took part in the attack. We are not told specifically that they entered secretly,
but they “took advantage of an unguarded position and created a disturbance inside the
castle” (Kuwata & Utagawa 1976, p. 262).
The Kōka-shū therefore carried out two successful raids, but the claim in Mikawa
Gofudo ki that they were “trained as shinobi” may simply be an interpretation placed upon
their successful service by an author who has already been influenced by the ninja myth
(Kuwata & Utagawa 1976, p. 262). Once again the use of mercenaries must also be ruled
out because a relationship typical of the times is suggested by the editors’ footnote that
explains that the Kōka men were jizamurai (Kuwata & Utagawa 1976, p. 245). The
members of the Kōka-shū were therefore in a similar position to that of many other small
landowners who enjoyed a relationship with a daimyō that was much looser than that which
would later be imposed upon them.
Kōka played a role in the Sekigahara campaign in 1600, when the leading Tokugawa
retainer in Ōmi Province was Yamaoka Dōami Kagetomo (1540-1603), who had rendered
great service to the Tokugawa at the time of Nobunaga’s death in 1582. Dōami destroyed
the strategic Seta bridge and thereby delayed the usurper’s march on Azuchi, and when
Ieyasu was trapped in Sakai Dōami and his brother helped him escape (Kuwata & Utagawa,

20

Journal of Global Initiatives

1976, p. 191). He had men of Kōka serving under him at his castle of Nagashima in 1600,
but the political situation was a complicated one because the Kōka area was the fief of
Minakuchi, and Minakuchi Okayama Castle was owned by Natsuka Masaie (1562-1600)
of the anti-Tokugawa faction, who also led Kōka men into battle. The result was that two
important sieges of the Sekigahara campaign saw men from Kōka serving on opposing
sides.
The two places were Takatori and Fushimi. Takatori came under attack from
Matsukura Shigemasa, who used troops from the anti-Tokugawa Minakuchi fief in his
operation. The Kōka-shū were sent to infiltrate the castle but were all wiped out except for
one man who was captured alive. His nose and ears were sliced off and his fingers and toes
severed. He was then sent back to the besiegers as an example of the treatment they might
expect (Kuwata & Utagawa, 1976, p. 197).
By contrast, the other castle at Fushimi had men from Kōka actively engaged in its
defense on the Tokugawa side in one of the most decisive actions of the Sekigahara
campaign. The castle was under the overall control of Torii Mototada and among the
garrison were over 100 samurai from Kōka under Yamaoka Kagemitsu, the younger
brother of Dōami, who were stationed in the castle bailey known as the Nagoya-maru. No
impression was made upon the fiercely defended fortress, so Natsuka Masaie seized
hostages from the defenders’ families in Kōka in the absence of their menfolk. An arrow
letter was loosed into the castle, informing the men that if they cooperated by setting fire
to Fushimi they would be richly rewarded. If they refused their wives and children would
be crucified. In order to save them the wretched men agreed, set fire to a tower, and took
down a section of the wall (Kuwata & Utagawa, 1976, p. 187). The enemy broke in and
after much desperate fighting Fushimi fell with a huge loss of life including all of the Kōka
contingent who had still remained loyal. They had however inflicted considerable losses
upon their enemies and also bought precious time that Ieyasu was able to exploit at
Sekigahara. Yamaoka Dōami then took his revenge by besieging Natsuka Masaie in
Minakuchi Castle. Having fled there from Sekigahara without firing a shot Masaie knew
his cause was hopeless and committed suicide. After the battle eighteen men from Kōka
who had been involved in the forced treachery and had then been captured alive were
crucified on Ieyasu’s orders (Kuwata & Utagawa, 1976, p. 160).
The sieges of Takatori and Fushimi present an interesting picture. First, there are men
from Kōka fighting on both sides and it is tempting to see this as evidence for mercenary
warfare, but this can be ruled out for the same reasons that were earlier applied to Iga: the
imposition of daimyō control and the effects of the Sword Hunt. Second, the picture
presented by their activities is a mixed one because at Takatori we see the classic ninja
skills of castle entry being carried out unsuccessfully, while at Fushimi the only secret
operation that is performed comes about because loyal samurai are forced to turn traitor.
The long service rendered to the Tokugawa by Yamaoka Dōami and the death of 100
men from Kōka outweighed the act of treachery by some of their number, so the
descendants of those who died at Fushimi were taken into Ieyasu’s service and Dōami was
placed in charge of them. He was granted 9,000 koku, 4,000 of which were for his Kōka
estates, and the next we hear of the Kōka men is that they are acting as guards at Edo castle
in a unit consisting of ten mounted samurai and 100 foot soldiers. These same guards served
at the siege of Osaka in 1614-1615, but in the capacity of a troop of musketeers, not shinobi.

Stephen Turnbull

21

Their appointment to Edo was a reward for good service, although it is conventionally
explained as a conscious act whereby Tokugawa Ieyasu took the supposedly powerful and
independent military cult of the Kōka ninja under his own wing. He then gained exclusive
use of them and thus effectively neutralized any capability a possible rival might have to
recruit ninja. I now believe that the situation was quite different and their appointment as
guards at Edo shows how firmly the Kōka-shū had been integrated into Tokugawa military
society rather than ever having being separated from it. This aspect of the ninja myth
therefore derives from their close relationship with the ruling Tokugawa family and not
from any notional independence of action.

Elite Warriors or a Warrior Elite?
The claim to exclusiveness by Iga and Kōka in relation to secret warfare is further
undermined by the fact that similar exploits are to be found throughout Japan, for whom
the second criterion states that such practitioners were not ordinary samurai but a skilled
elite. In the historical accounts those who carry them out may be described as kanchō
(spies) or teisatsu (scouts), or the emphasis may simply be on the tasks they are performing
using the word shinobi as in shinobi itte (entered in secret). One surprising example of the
latter involves no less a person than Toyotomi Hideyoshi. In 1579, when the future ruler
of Japan was still only one of Oda Nobunaga’s generals, the Shinchō-Kō ki records how,
“in the middle of the night, Hashiba Chikuzen no Kami Hideyoshi stole into (shinobi itte)
the fort of Kaizōji in Harima Province and seized it” (Kuwata, 1965, p. 249; Elisonas &
Lamers, 2011, p. 314).
Both the Hōjō of Odawara and their rivals the Satake would send foot soldiers to spy
on the enemy lines. They are referred to sometimes as shinobi but also as kusa (grass)
because of their technique of hiding in the long grass for hours on end, often returning the
following day to report. The Hōjō Godai ki contains an account of how kusa used in this
way by the Satake tried to intercept the Hōjō’s mounted scouts as they returned from a
reconnaissance mission (Hagiwara, 1966, pp. 395-397). The Hōjō also used men known as
rappa, a compound that blends the characters for war or disorder with that for a wave as in
the word tsunami, and it must have felt like a “wave of disorder” in 1580 when they entered
secretly (shinobi no iri) night after night into Takeda Katsuyori’s lines to cut through the
tethers of horses and generally cause havoc. The rappa originated from local mountain
bandits or pirates but were fully integrated into the army of Hōjō Ujinao because we are
told that there were 200 of them who operated in four squads and that they received fuchi
(stipends) (Hagiwara, 1966, pp. 397-398).
The Ōu Eikei Gunki of 1698, which covers the civil wars in Tōhoku, contains a story
of a man who is genuinely skilled in shinobi activities and appears in one of the best
accounts of such operations. The incident occurred during the siege of the castle of Hataya
in 1600, where he performed a clever piece of psychological warfare:
There was someone inside Hataya castle with renowned shinobi skills (shinobi no jōzu),
and that night he entered the enemy camp secretly (shinobi iri). He took a ban sashimono
(company flag) from Naoe Kanetsugu and a fukinuki (streamer) from the barracks of

22

Journal of Global Initiatives

Kurogane Magoza’emon, returned and hung them from a high point on the front gate of
the castle. (Imamura, 2005, p. 815)

The word shinobi appears twice in the above account to describe the man’s skills and
his activity, but he has no connection with Iga or Kōka, nor is there any suggestion that he
belonged to a specialized elite. He was just “someone” who had these skills.
Higo province provides a further illustration. It was cited earlier in this article as a
place that was similar to Iga in many ways because its independently-minded barons
formed shū to fight for neighboring daimyō when it suited them. Their independence came
to an end in 1587 at the hands of Toyotomi Hideyoshi in much the same way as Iga had
been pacified by Nobunaga. During their final acts of rebellion we come across several
incidents which, if they had been performed by men from Iga, would have been credited to
ninja. Supply columns were harassed and castles were attacked (Araki, 1987, p. 191),
culminating in a siege of the rebel stronghold of Tanaka Castle (Araki, 1987, pp. 92-93;
Ōyama, 2003, pp. 124-126). As the siege of the tiny castle wore on Hideyoshi’s frustrated
commander used an arrow to convey a message to Hebaru Chikayuki, the one member of
the castle command whom he believed to be susceptible to negotiation. The ploy
succeeded, and Chikayuki persuaded a retainer called Usono Kurandō to kill the castle
commander Wani Chikazane, against whom Kurandō bore a grudge (Kumamoto City,
2000, p. 431). The Wani Gundan reads:
Kurandō agreed and in the middle of the night of the 6th day of the 12th month he entered
secretly (shinobi no itte) into Chikazane’s private quarters and killed him, cutting off his
head he escaped into the enemy lines. Chikayuki sent a signal by means of a beacon and
the attack was launched. At the same time Hebaru Chikayuki set fire to the inner bailey
and because of a strong wind the castle was engulfed in flames. (Kumamoto City, 2000, p.
65)

Unrest spread to the Amakusa Islands two years later and in October 1589 Konishi
Yukinaga dispatched 3,000 troops against the Christian rebel baron Juan Shiki Rinsen
(Tsuruta 1981, pp. 49-50). The force advanced without hindrance as far as the island of
Shimojima, unaware that they were being led into a trap. There Rinsen caught them in a
surprise attack and massacred them (Araki, 1987, p. 138). Yukinaga then led a much larger
army against them, so Rinsen sent a secret envoy to Yukinaga’s ally Ōyano Tanemoto to
discuss peace terms. The man is referred to using the two character compound shinobi
mono or ninja, which obviously indicates no more than that he was on a covert mission
because his intention was to discuss peace (Tsuruta, 1981, pp. 50 & 58). So here we have
a series of military actions involving subterfuge, secret entry, assassination, ambush, and
covert operations without a mercenary from Iga or Kōka in sight, and it would also appear
that the Higo Rebellion provides instances of fighting men who are not ninja and ninja who
are not fighting men!
The conclusion must therefore be that the practitioners of secret warfare in these
historical accounts were warriors who operated within the usual command structure of a
daimyō’s army. They were elite warriors but not a cult-like warrior elite from a distant
province who served as mercenaries. The other popular notion that the secret warfare skills
displayed in these provinces was provided by refugees from Iga may also be dismissed.

Stephen Turnbull

23

Apart from instances where “Iga-shū or Kōka-mono” are used as a casual and inaccurate
shorthand for shinobi no mono there is no suggestion of any connection.
The other important point that the above accounts have in common is that when the
word shinobi is used it is intended as an adverb, not a noun. It indicates how something is
carried out, not who does it, and the subsequent shift in meaning from adverb to noun
represents an important element in the invented tradition of the ninja of Iga and Kōka.
When a flag is stolen by someone at Hataya he is praised for what he does. When a flag is
stolen by someone from Iga he is praised for what he is: a ninja. The word shinobi also
appears as often in these accounts as it does for Iga and Kōka, if not more. Just as these
areas were no different politically from anywhere else run by kunishū, neither were they
any different militarily from any other province, including those run by daimyō, in the
frequency of their employment of secret warfare.
Yet this is not the whole story of the growth of the Iga-Kōka ninja myth and the
development of the modern cult. There also appears to have been an appropriation of other
people’s actions, of which the most blatant example occurs during the final act of rebellion
against the Tokugawa, where we find the last ever descriptions of secret operations. They
took place during the largely Christian Shimabara Rebellion of 1637 to 1638, which
finished with the siege of the castle of Hara (Fujita, 2012, pp. 48-51). The operation against
Hara receives detailed coverage by Yamaguchi, who makes use of material put together
many years later by Ukai Shōsan. He is said to be a descendant of Ukai Kanemon, one of
ten men in a Kōka contingent of 100 taken to Hara by Matsudaira Nobutsuna (1596-1662)
as reinforcements for the ineffective besiegers (Yamaguchi 1969, p. 59). Ukai Shōsan’s
work is presented in the form of a diary that relates how on the 6th day of the 1st month
ninja from Kōka secretly survey the defenses of the castle “from the Arima lines to the
second bailey” and the results are reported back to Edo (Yamaguchi 1969, pp. 59 & 255).
On the 21st day of the same month a further raid is carried out, “They raided from the
Kuroda lines near the Western beach and worked together to capture thirteen bags of
provisions which were the enemy’s only means of survival. That night they again infiltrated
the enemy castle and acquired secret passwords.” (Yamaguchi 1969, pp.60 & 255-256).
On the 27th day another raid is carried out from the Hosokawa lines which results in the
capture of an enemy flag. Two named ninja are wounded and are out of action for forty
days. By the 20th day of the 2nd month it is known that the garrison have been reduced to
eating only seaweed and grass and a counter-attack by them is driven back. From the 27th
day until the fall of the castle raids are launched, taking the third and second baileys.
(Yamaguchi 1969, pp. 60 & 256).

Yamaguchi then goes back to describe an earlier raid. The approach of the ninja is
covered by the Hosokawa troops firing a volley. This alerts the garrison who obligingly
plunge the castle into darkness by extinguishing the lanterns. The ninja wait for some time
and eventually enter, although one of the ninja falls into a pit and has to be rescued by his
comrade. They are discovered when the pine torches are re-lit, so they take one of the flags
bearing a Christian cross and make their escape. A hail of stones follows them as they
withdraw (Yamaguchi 1969, pp. 62 & 258-259).
All this is very convincing, but the picture is not all that it seems, because Hosokawa
Tadatoshi (1586-1641), the daimyō of Kumamoto, also used secret warfare at Hara, and
the incidents described in the Hosokawa family records appear to be the same ones. In the

24

Journal of Global Initiatives

Hosokawa account the raid of the 27th day of the 1st month is carried out not by a ninja
from Kōka but by a retainer of the Hosokawa identified as a kogashira (“lieutenant”) in
the company of Hirano Jibuzaemon. He infiltrates (shinobi iri) the castle by night and steals
a flag, for which exploit he is awarded five silver coins. The following night another
individual who is a shinobi no mono in the company of Yoshida Suke’emon returns from
a raid with rebel heads, but finding that he is being pursued by 14 or 15 men discards all
but one of his trophies (Toda, 1988, p. 191). On the 13th day of the 2nd month two shinobi
no mono sneak into the castle in a very unusual manner because they have ropes attached
to them so that if they are shot they can be pulled back, but they return successfully with
information. The account then tells us that the Hosokawa had ten shinobi no mono within
their ranks and that they were used night after night. One is named as the son of the karō
(senior retainer) Sado; another is Uehara Heinosuke, who hears the striking of a bell as he
enters the castle. Realizing that it means that the guard is being changed, he conceals
himself (Toda, 1988, p. 199). It is also interesting to note how narrow is the definition of a
shinobi skill in the Hosokawa archives. Digging a hole under the wall to provide entry to
the castle (a technique used by the Hosokawa at Hara) is not a shinobi operation even
though it provides exciting reading with the defenders thrusting spears down into the gap
to flush out infiltrators. Shinobi is applied only to entering a castle in secret by climbing its
walls, not tunnelingtunnelling beneath them (Toda, 1998, pp. 198-199).
The Hosokawa records present a dispassionate and disinterested account of secret
warfare by skilled individuals that fits in very well with other accounts such as the flagstealer of Hataya, but the Kōka exploits are difficult to believe because they are based on
the assumption that Matsudaira Nobutsuna called in at Minakuchi Castle in Kōka and
recruited 100 battle-ready ninja. The Kōka-shū in Edo Castle were trained for wartime use
as musketeers, not ninja, and it is difficult to envisage a large group of ninja being retained
in peacetime at Minakuchi for possible use as a rapid deployment squad. Falling into a hole
also indicates that they were not highly trained ninja. Instead Matsudaira Nobutsuna took
troops from Kōka simply because it was a fief directly held by the Tokugawa family, and
as gaining control of Hara Castle was the object of the entire operation some of the Kōka
men may have volunteered for shinobi duties. A desire to emulate the deeds of their
ancestors, a common obsession among samurai, would be a sufficient explanation for their
enthusiasm. My conclusion is therefore that men from Kōka were involved in the siege of
Hara but that Ukai Shōsan padded out their brave yet modest achievements by borrowing
material from the archives of the Hosokawa family. So instead of independently confirming
the authenticity of ninja the Ukai material is the best argument so far for the ninja cult being
an invented tradition.

Conclusion
I therefore conclude that the authenticity of the Iga Kōka ninja tradition fails against all
three of the suggested criteria in terms of the existence of secret warfare, the elite
practitioners of it, and its narrow geographical location. Undercover operations were
performed throughout Japanese history but were carried out by skilled warriors who did
not belong to a hereditary tradition. Iga and Kōka did not have a monopoly, nor is there
any evidence for a transmission from there to other provinces after 1581. In fact the
opposite is true. If the presence of shinobi in an account indicates that secret warfare is

Stephen Turnbull

25

taking place, then Iga and Kōka are remarkable not for how many references there are to
them but for how few. The ninja of Iga and Kōka therefore present a classic example of an
invented tradition in terms of Hobsbawm and Ranger’s definition. There is a deliberate
“attempt to establish continuity with a suitable historic past” whereby ancient records have
been re-interpreted and exaggerated to reinforce a highly localized understanding of a
military phenomenon (Hobsbawn and Ranger 1983, p. 1). Hobsbawn and Ranger (1983)
also identified the “use of ancient materials to construct invented traditions of a novel type
for quite novel purposes” (p. 4). This seems to be what has happened at Shimabara.
I do not however believe that the Iga-Kōka ninja myth or the modern cult that
developed from it represent a total fabrication. All invented traditions have a basis in fact,
no matter how tenuously the links may be made between the developed tradition and
recorded history. In Iga and Kōka there must have been some genuine belief in a unique
local expertise that was bolstered by folk memories and old soldiers’ tales, and the best that
can be said for their plagiarism of other people’s exploits is that it supports one great ninja
stereotype: they were very good at stealing things! Yet even if the Iga-Kōka ninja cult
draws upon little more than the manipulation of folk memories and historical records, any
tradition that takes shape in about 1620 and continues to the present day is worthy of more
attention and respect as a cultural property than is commonly given to other aspects of the
samurai tradition. As the Iga and Kōka ninja tradition is older than the 47 Rōnin and even
predates bushidō it should not be dismissed but celebrated as Japan’s oldest martial
invention and, through its modern cult-like manifestation, as Japan’s greatest martial
fantasy.

References
Araki, E. (1987). Higo Kunishū ikki. Kumamoto, Japan: Kumamoto Shuppan Bunka Kaikan.
Asawaka, K. (1929). The Documents of Iriki: Illustrative of the Development of the Feudal
Institutions of Japan. New Haven, CT: Yale University Press.
Aston, W. G. (1972). Nihongi: Chronicles of Japan from the Earliest Times to A.D. 697 (Reprint).
City, VT: Tuttle.
Blair, E. H. & Robertson, J. A. (Eds.). (1903). The Philippine Islands, 1493-1803 (Volume 1, Part
24.). Cleveland, OH: Arthur H. Clark Co.
Elisonas, J. (1991). Christianity and the daimyo. In J. W. Hall & J. L. McLain (Eds.), The Cambridge
History of Japan. Vol. 4 Early modern Japan (pp. 301-372). Cambridge, MA: Cambridge
University Press.
Elisonas, J. S. A. & Lamers, J. P. (Trans. & Eds.). (2011). The Chronicle of Lord Nobunaga by Ōta
Gyūichi. Leiden, Netherlands: Brill.
Ferejohn, J. A. & Rosenbluth, F. M. (Eds.). (2010). War and State Building in Medieval Japan.
Stanford, CA: Stanford University Press.
Fujiki, H. (2005). Katanagari. Tokyo, Japan: Iwanami Shoten.
Fujita, K. (2012). ‘Kōga ninja’ no jitsuzō. Tokyo, Japan: Yoshikawa Kōbunkan.
Hagiwara, T. (Ed.). (1966). Hōjō Shiryō Sōshū in Sengoku Shiryō Sōsho (Vol. 21). Tokyo, Japan:
Jinbutsu Ōraisha.
Hall, J. W. (Ed.). (1981). Japan Before Tokugawa: Political Consolidation and Economic Growth,
1500 to 1650. Princeton, NJ: Princeton University Press.
Hayashi, S. (1954). Shimabara Hantō-shi (Vol. II). Nagasaki, Japan: Minami Takaki-gun Board of
Education.

26

Journal of Global Initiatives

Hobsbawm, E. & Ranger, T. (Eds.). (1983). The Invention of Tradition. Cambridge, MA: Cambridge
University Press.
Imamura, Y. (Ed.). (2005). Ōu Eikei Gunki. Tokyo, Japan: Jinbutsu Ōraisha.
Konstam, A. (2011). Pirate: The Golden Age. Oxford, England: Osprey Publishing.
Kumamoto City. (2000). Kumamoto-shi shi kankei shiryōshū (Vol. 4: Higo koki shūran). Kumamoto
City, Japan: Kumamoto City.
Kuwata, T. (Ed.). (1965). Shinchō-Kō ki. Tokyo, Japan: Jinbutsu Ōraisha.
Kuwata, T. & Utagawa, T. (Eds.). (1976). Kaisei Mikawa Gofudo ki. Tokyo, Japan: Akita Shoten.
Lamers, J. P. (2000). Japonius Tyrannus: The Japanese Warlord Oda Nobunaga Reconsidered.
Leiden, Netherlands: Hotei Publishing.
McCullough, H. C. (1959). The Taiheiki: A Chronicle of Medieval Japan. New York, NY: Columbia
University Press.
Nitobe, I. (1905). Bushido: The Soul of Japan: Tenth Revised Edition. New York, NY: Putnam’s.
Ōyama, R. (2003). Hideyoshi to Higo kunishū ikki. Fukuoka, Japan: Kaichosha.
Park, Y. (1978). Admiral Yi and his Turtleboat Armada. Seoul, South Korea: Hanjin Publishing
Company.
Sasama, Y. (1968). Buke Senjin Sakuhō Shūsei. Tokyo, Japan: Yūzankaku.
Shimizu, N. (2009). Sengoku ninja wa rekishi o dō ugokashita no ka? Tokyo, Japan: Best Shinsho.
Souyri, P. (2010). Autonomy and War in the Sixteenth Century Iga Region and the Birth of the Ninja
Phenomena. In J. A. Ferejohn & F. M. Rosenbluth (Eds.), War and State Building in Medieval
Japan (pp. 110-123). Stanford, CA: Stanford University Press.
Sugawara, M. (1985). Lives of Master Swordsmen. Tokyo, Japan: The East Publications.
Sugiyama, H. (1974). Sengoku Daimyo. Tokyo, Japan: Chūō Kōronsha.
Toda, Toshio (1988). Amakusa-Shimabara no ran: Hosokawa han shiryō ni yoru. (Tokyo: Shin
Jinbutsu Ōraisha).
Turnbull, S. (1991). Ninja: The True Story of Japan's Secret Warrior Cult, London, England: Firebird
Books.
Turnbull, S. (2011). The Revenge of the 47 Ronin: Edo 1703. Oxford, England: Osprey Publishing.
Tsuruta, S. (1981). Tenshō Amakusa Kassen no kōsatsu. Kumamoto Shigaku, 55, pp. 47-58.
Urban, W. (2006). Medieval Mercenaries: The Business of War. London, England: Greenhill Books.
Yamaguchi, M. (1969). Ninja no Seikatsu. Tokyo, Japan: Yūzankaku.
Yokoyama, T. (1992). Nobunaga to Ise. Iga - Mie Sengoku monogatari Tokyo, Japan: Sōgensha.
Yokoyama, T. (2006). Iga Tenshō no ran. Tokyo, Japan: Shinpu Shobō.
Yoshida, Y. (Ed.). (1979). Taikō ki (Vol. III). Tokyo, Japan: Kyōikusha.
Yūki, S. (1988). Kōka jōkaku gun. Rekishi Dokuhon, 482 114-121.