What Did the Kōyō Gunkan Mean by a “True Battle”?

A fresh translation

Original text

Words and concepts

• 本の合戦 (hon no kassen)
  Literally a “true,” “real,” or “genuine battle.” Here hon marks an approved category, not simply an event that actually happened.
• 国持
  A lord who holds a province or substantial domain. The definition begins with the status and scale of the commanders rather than with the experience of individual fighters.
• 二三万 / 二千三千
  “Twenty or thirty thousand” and “two or three thousand.” The contrast establishes an order of magnitude; it should not be converted into an exact modern count.
• 実否をつける
  To bring the question to a definite result or settle the matter in fact. The phrase helps distinguish a decisive test from an advantage attributed to surprise or obstruction.

A category before a count

The passage does not begin by asking how many battles occurred. It first decides which engagements deserve a privileged name. That difference matters. A list of events tries to include every qualifying occurrence; a category such as “true battle” ranks some occurrences above others. Chapter Fifty is therefore giving us an argument about value as much as a report about warfare.

Its requirements accumulate quickly. The opponents must be territorial lords. Their forces must be large. Each side may receive outside aid, but each must remain under one principal commander. The action must occur in daylight. Neither side is to rely on a moat, river, palisade, or sudden betrayal. The armies must close and test themselves directly. Only then does the writer grant the label hon no kassen.

No single requirement is offered as sufficient. A daytime action can still be excluded because a defended barrier shaped it. A large operation can still fall short if surprise decided it. A hard-fought local engagement can still be denied the status of a “great battle” because its forces numbered only in the thousands. The definition works by narrowing the field until very few examples remain.

The imagined equality of the test

The ideal resembles a controlled comparison. Both sides bring armies on the same broad scale, each answers to one commander, and conspicuous external advantages are stripped away. Daylight removes the cover of night. The rejection of moats, rivers, and palisades removes protective obstacles. The rejection of betrayal removes a sudden change of allegiance. What remains is supposed to reveal the quality of the opposing forces through direct action.

That is why “true” should not be mistaken for “historically real.” The chapter does not claim that night attacks, sieges, river crossings, ambushes, or defections were imaginary. Just before the selected lines, it discusses victories achieved through surprise and through an opponent’s lack of caution. Such events were real enough to require explanation. The writer’s point is that they do not provide the kind of evidence he wants when comparing martial strength.

The category is thus argumentative. It decides in advance which causes will count as a fair measure and which will be treated as interference. Terrain and fortification become obstacles to comparison rather than parts of skill. Surprise becomes a reason to discount an outcome rather than a form of successful planning. The definition does not merely describe battle; it edits battle into an ideal test.

Why exclude methods that often decide wars?

From the standpoint of a campaign, the exclusions are striking. Commanders ordinarily care about gaining an advantage before direct collision. They use roads, rivers, fortified positions, timing, intelligence, alliances, and deception because those things affect survival and success. A definition that removes them does not tell us how war normally worked. It tells us what this chapter wants a victory to prove.

A surprise victory may demonstrate preparation, secrecy, or an opponent’s error, but it leaves room to say that the losing army was never tested under equal conditions. A victory behind a palisade may demonstrate intelligent choice of ground, but the barrier complicates a simple comparison of courage and cohesion. Betrayal makes the outcome even harder to attribute to the two commanders and their original forces. Chapter Fifty excludes these factors so that the result can be narrated as a verdict on the armies themselves.

This also explains the emphasis on spears. The wording does not present a technical formation or teach a particular thrust. “Meeting spear to spear” stands for visible, reciprocal engagement. Each side is imagined as able to answer the other in the same arena. The phrase supplies an image of directness, not a complete inventory of the weapons used.

Two battles selected for Shingen

Once the criteria are established, the writer names Kawanakajima and Mikatagahara and declares Takeda Shingen victorious in both. In the sentence immediately preceding the excerpt, the Kawanakajima example is identified with Eiroku 4. The point is not merely chronological. These two engagements are presented as the rare cases that pass the chapter’s demanding test.

Selection and praise reinforce one another. The definition elevates the kind of battle in which the work says Shingen excelled; the named examples then make the definition look historically grounded. A reader should notice that circular movement. The chapter is not a neutral referee applying rules created elsewhere. It belongs to a textual tradition centered on the Takeda house, and its taxonomy helps organize remembered warfare around Shingen’s reputation.

That observation does not require dismissing the passage. On the contrary, it makes the passage more useful. If we read its verdicts as transparent scorekeeping, we obtain little more than a claim that Shingen won. If we read how the category is constructed, we see the standards by which the writer wanted martial reputation to be judged: scale, openness, unified command, direct collision, and a result not easily explained away by an asymmetric advantage.

The numbers are thresholds, not a census

The contrast between twenty or thirty thousand and two or three thousand gives the definition scale. Yet the text does not provide a muster roll, explain whether every attendant is included, or distinguish combatants from the wider body accompanying an army. The numerals belong to the rhetoric of classification. They separate the engagements that attract broad notice from smaller contests said to occur throughout the provinces.

For that reason, the figures should remain approximate in translation. Writing “20,000–30,000” may look more precise on a modern page than the source’s compact ni-san-man warrants. The same caution applies to “two or three thousand.” Chapter Fifty is drawing a boundary by magnitude, not offering a dataset from which exact totals can be reconstructed.

The sentence also reveals whose warfare defines greatness. An engagement involving several thousand people could be catastrophic to those present and decisive for a locality. It nevertheless falls outside the chapter’s “great battle” category. The viewpoint is that of province-holding lords and public reputation. The experiences of smaller forces are subordinated to the scale at which major houses compare power.

What the passage does not prove

These lines do not prove the actual strength of every army at Kawanakajima or Mikatagahara. They state the scale required by the chapter’s definition and then name examples. Independent reconstruction would require other witnesses, documents, and judgments about what different counts represent.

They do not prove that the outcomes were understood identically by all participants. The declaration of two Shingen victories is the Kōyō Gunkan’s verdict. It is evidence for the work’s representation of the battles, not a substitute for comparing accounts from other sides.

They also do not establish a timeless Japanese code of fair battle. The passage belongs to one work and one argumentative setting. Its rejection of barriers and surprise sits awkwardly beside many celebrated episodes of siegecraft, ambush, night attack, and maneuver. That tension is precisely why the definition should be read as a deliberate ideal rather than an automatic rule followed by every warrior.

Finally, “true battle” is not a license to reduce history to a contest of bravery. The chapter itself must remove numerous factors before it can produce that kind of test. Real campaigns did not remove them. Logistics, terrain, intelligence, political relationships, fortification, and timing remained part of what made action possible.

A definition that exposes the narrator

Historical writing often becomes clearest where it sorts events. Words such as “great,” “decisive,” “proper,” and “true” appear to describe an object, but they also expose the observer’s scale of judgment. Here the boundary around hon no kassen reveals what the narrator wants an outcome to say about a commander and his house.

The value of these three statements therefore lies beyond the two battles they name. They show military memory being arranged: smaller actions are pushed aside, asymmetric victories are discounted, direct confrontation is privileged, and Shingen is placed at the center of the surviving examples. The category is not a window without glass. It is the glass—shaped, colored, and worth examining.

Source

Kōyō Gunkan (甲陽軍鑑), Chapter Fifty (品第五十), passage defining hon no kassen (本の合戦). The quoted text is checked against the public Wikisource transcription of Chapter Fifty, which is based on the 1893 Onkodō edition, and its linked National Diet Library image, frame 227. For an earlier printed witness and bibliographic context, the NIJL/CODH record describes a 1656 Kyoto edition attributed on the item to Kōsaka Danjō, comprising twenty-three booklets and twenty scrolls.