Giles and Griffith: how to choose a translation of Sun Tzu

Choosing a translation requires looking beyond the name on the cover. The treatment of terms, the notes and the editorial interventions make it possible to compare Giles and Griffith according to how the book will be used.

Finding the best translation of Sun Tzu requires being clear about why you are going to read it.manuscripts and translations of Sun Tzu sets out the different versions; here, we compare specific decisions by Lionel Giles and Samuel B. Griffith.

The comparison starts from Giles's annotated English translation of 1910 and Griffith's edition published in 1963. The Spanish explanations of the terms are glosses for this analysis, not quotations from a commercial Spanish edition.

The text and the explanations that accompany it

In Giles, the notes present interpretations by Chinese commentators and explain translation decisions. The reader can follow a passage and then examine why it admits a particular reading. This apparatus is especially useful when a Chinese word allows several interpretations.

Griffith's edition includes a historical introduction, comments, appendices, a bibliography and an index, as well as other editorial material. These components make it possible to distinguish the study that accompanies the treatise from the translated chapters.

This difference in organisation should be judged according to the task. To follow an argument without interruption, it matters being able to see where the main text continues. To study a word, it matters finding the translator's explanation and knowing whether alternatives are considered. Ease of reading and the ability to check an interpretation are not the same criterion.

Nor is it enough to set up a contrast between 'philologist' and 'military man' as though they were mutually exclusive categories. A useful comparison asks what reasoning each note offers, what problem it is trying to solve and how much the reader can verify. The vocabulary of a translation may sound more technical without that, in itself, guaranteeing greater accuracy in relation to the original.

Three decisions that make it possible to compare the versions

The title of the first chapter: planning or evaluating

Giles usesLaying Plans, whereas Griffith usesEstimates. The two choices guide the first impression of the chapter in different ways.

As a reading guide, 'making plans' invites one to think about what will be done, whereas 'estimates' directs attention to the conditions that must be examined first. This difference in emphasis does not establish an absolute opposition: estimating can be part of planning. The useful thing is to see how the title fits with the chapter's development, without deciding solely from the isolated impression of a single word.

Dao: moral law or moral influence

For one of the opening factors, Giles usesMoral Law, whereas Griffith opts for a formulation related to moral influence.

The Spanish glosses 'ley moral' and 'influencia moral' suggest slightly different questions. The first may make one think of a principle; the second, of the effect of behaviour on other people. It is a comparison of the emphasis of the expressions, not a claim that the Chinese term has two completely separate meanings.

To assess the passage, its later explanation must also be read: the relationship between those who lead and those who follow them. Translating an isolated word can hide what the following sentence clarifies. That is why it is worth comparing complete units of meaning and not building a classification of translations using a dictionary of loose equivalents.

Fa: method and discipline or doctrine

For the fifth factor, Giles usesMethod and discipline, whereas Griffith choosesdoctrine. The difference shows how the same concept can be presented through a descriptive expression or through a more condensed term.

'Method and discipline' breaks the concept down into two relatively common words. 'Doctrine' condenses it into a category that the reader must understand within the military context. The second formulation should not automatically be taken to mean a religious belief; the first does not authorise us to reduce the passage to individual obedience either.

Editorial interventions also matter. In certain places, Griffith notes in his edition that he has reorganised material or adopted decisions about the layout of the text. When an edition makes this kind of intervention, the comparison cannot be limited to looking for the same paragraph number.

If two editions do not present exactly the same order, the content must be located and the notes explaining the change must be read. A translation may make questionable decisions and, at the same time, offer the reader enough information to recognise and examine them.

How to choose an edition according to use

For a study focused on language problems, it is worth consulting Giles's annotated edition and working through each passage alongside its notes. For a reading that includes an extensive historical and military presentation, Griffith's organisation offers that route. These guidelines relate to the components of each edition, not to any universal superiority of one translation over the other.

A newcomer can do a simple test: read the same extract in both, explain in their own words what they understand, and then review the notes. The doubts that survive that second reading matter more than the first impression of fluency. An apparently transparent text may have quietly resolved an ambiguity that it is important to know about.

It is also necessary to distinguish between translation and critical apparatus. Two editions may offer similar wording of the text and yet provide very different experiences because one includes extensive notes and another removes almost all commentary. If the aim is to investigate a specific expression, the notes may be just as relevant as the fluency of the translation.

In Spanish, the check should begin with the credits. You need to identify who produced the Spanish version, what text it is based on and whether it preserves introductions and comments. The name of Giles or Griffith on the cover is not enough to attribute a Castilian sentence to them literally. Nor should it be assumed that a reprint retains all the material from the original edition.

This point is especially important when a Spanish edition says it is based on an English translation. There may be a second translation decision between English and Spanish. A Castilian word should not be attributed directly to Sun Tzu, Giles or Griffith without checking who actually chose it.

The historical comparison can be expanded withAmiot's translation; the shift in interpretative framework, withThomas Cleary's interpretation. It is worth keeping these questions separate: reception history, word choice and philosophical reading are not interchangeable criteria.

The original editions compared here pre-date the discovery ofthe Yinqueshan bamboo manuscripts in 1972. Studying those materials requires later documentation; a recent date on the cover of a reprint does not prove that its content has been updated.

For that reason, choosing between Giles and Griffith depends on what you want to check. The reader interested in comparing vocabulary can work with parallel passages and notes. Anyone looking for historical context can also pay attention to introductions and appendices. In both cases, identifying the translator's interventions avoids confusing an editorial interpretation with an inevitable equivalent of the Chinese text.

Before choosing an edition, review a sample that includes text and notes. Check the translation credits and which materials it preserves; the number of pages alone does not answer those questions.

View all articles in Sun Tzu translations: manuscripts, versions and reading criteria