Concerning the Nanjing Rules of Interpretation

The Nanjing rules are based on nineteen examples of I Ching divination recorded in the Zuo Commentary, a record of the Spring and Autumn period (722-464 B.C.).  Three similar examples from the same period are contained in Guoyu, “State Speeches,” written and compiled at a later date.  As for the rules, I will confess at the outset that I do not think much of them.  Here is my analysis, based on the translation of the text and the version of the rules contained in Zhouyi:  A New Translation with Commentary of the Book of Changes by Richard Rutt.  I shall summarize the information from the divination examples on which the rules are based, examine the rules in light of this information and offer a simplified version that still fully accounts for all the data, explain my objections to the rules, and offer an alternative view.


I.  The data

Of the 22 examples in the Zuo Commentary and Guoyu, 16 specifically describe results that were obtained by yarrow stalk divination.  The other six merely quote the I Ching and draw conclusions therefrom (although the numbers in three of the quotes still follow the rules; the other three are not specific enough for the rules to apply).  It is the 16 divinations that will be the focus of attention here.

(1) In ten of the examples (1, 2, 3, 5, 6, 11, 14, 15, 16, and 19), the result is identical in form to the first one:  “the line of Guan (Hexagram 20) that is changed to make Pi (Hexagram 12)” (in other words, line 4 of Hexagram 20, or 20:4 for short).  All of these use the single indicated line statement as the oracle; and all but three additionally make at least indirect reference to the new hexagram that results from changing the line.

(2) In two of the examples (4 and 9), the result is a single hexagram.  The hexagram statement is the oracle in each case.

(3) In example 21, the result is “an 8 of Tai (Hexagram 11).”  (More about this later.)  The hexagram statement is again the oracle.

(4) In two more examples (10 and 20), an 8 is part of the result:  “an 8 in Gen (Hexagram 52)” and “Zhun (Hexagram 3) turning to Yu (Hexagram 16) by an 8.”  Both of these cases require multiple moving lines for the transformations, but hexagram statements are used as the oracle in each case.

(5) In example 22, the result is “Qian (Hexagram 1) changing to Pi (Hexagram 12).”  This requires three changing lines; both hexagram statements are used as the oracle.


II.  The original rules

Remainder
from 55
Line of
hexagram
 671819 6 (top)
 5817  5
 4916  4
 31015  3
 21114  2
 11213  1 (bottom)

The scholars who came up with the Nanjing rules set out to account for why some of the divinations resulted in just one line, and some used one or both hexagram statements.  Their system begins by assigning conventional line numbers (6, 7, 8, 9) to all the lines in the hexagrams used in the examples.  Then, for each example, the line numbers are added together, and the sum subtracted from 55 (which happens to be the sum of numbers 1-10).  The remainder (1-19) is used as in the accompanying table to indicate one of the six lines.  The oracle is then read as follows.  My comments are in brackets.

(1) If the base hexagram is composed entirely of stable lines, its hexagram statement is the oracle.  The indicated line is irrelevant.
[Based on two examples:  4 and 9, the single-hexagram ones above.  This rule is subsumed by rule (4) (a) and does not need to be stated as a separate rule.]
(2) If the hexagram is composed entirely of changeable lines, it must be converted to the new hexagram and the hexagram statement of the new hexagram is the oracle, saving that in Hexagrams 1 and 2 the 7th line statement is used.  (There is no example of this in the Zuo Commentary.  The rule is extrapolated from the other rules.)
[Rule (3) already applies to this situation, as the indicated line will always be a changeable line.  I see no justification for a separate rule in the absence of any data.]
(3) If the indicated line is a changeable line, that line statement in the base hexagram is the oracle.  No notice is taken of other changeable lines.
[Based on ten examples:  the single-line ones above.  But note well:  there is more to the rule than this.  As discussed below, to come up with the rules, both changing and unchanging line numbers must be assigned to the various lines; but in this case, the indicated line is the only one allowed to change.  Plus, a new hexagram is formed based on the single changed line.]
(4) If the indicated line is stable, a hexagram statment is the oracle, chosen as follows:
(a) when the hexagram contains fewer than three changeable lines, the base hexagram statement is the oracle;
[Based on three examples:  21, plus the same examples as rule (1) above; this rule can replace the former.  All have zero or one changing line.]
(b) when the hexagram contains three changeable lines, they are all changed to make the new hexagram, and both hexagram statements together form the oracle;
[Based on two examples:  20 and 22.]
(c) when the hexagram contains four or five changeable lines, the new hexagram is formed as in (b) and the new hexagram statement is the oracle.
[Based on one example:  10, which has five changing lines.  Given the paucity of examples for these last three rules, and the fact that the numbers of lines represented are only zero, one, three, and five, the rules are equivalent if we make them symmetrical by applying the first to 0-1, the second to 2-3, and the third to 4-5 changing lines.]

III.  The simplified rules

These rules produce results identical to the ones above for all of the examples in the texts.

(1) If the indicated line is a changeable line, that line statement in the base hexagram is the oracle.  The indicated line is the only one allowed to change, and a new hexagram is formed based on the single changed line.
(2) If the indicated line is stable, a hexagram statment is the oracle, chosen as follows:
(a) when the hexagram contains zero or one changeable line, the base hexagram statement is the oracle;
(b) when the hexagram contains two or three changeable lines, they are all changed to make a new hexagram, and both hexagram statements together form the oracle;
(c) when the hexagram contains four or five changeable lines, a new hexagram is formed as in (b), and the new hexagram statement is the oracle.

IV.  Objections to the rules

My biggest objection to the rules stems from the fact that in twelve of the examples, the divination result is a hexagram with zero or one changing line whose interpretation is straightforward and needs no special rules.  The Nanjing rules are actually based on, and required to explain, only four of the 16 examples in the texts (10, 20, 21, and 22).  Furthermore, three of these are the ones that come from Guoyu, not Zuo, and their rules may be completely unrelated.  (There is an actual reason to believe this, as described below.)  This is a very small amount of data.

My next objection is that the numerologic system is contrived based on arbitrary assignment of line numbers to the hexagram lines.  To illustrate:  the divination result of the second example in the text is “the line of Zhun (Hexagram 3) that is changed to make Bi (Hexagram 8),” or 3:1.  Hexagram 3 with a changing line in the first position would normally be represented by lines 988878.  The sum of these numbers is 48; 55 - 48 = 7, which according to the rules indicates the top line as the result.  But we want the first line to be the result.  What to do?  Change the numbers to 966678.  This is still Hexagram 3, with a changing line in the first position; and 55 - 42 = 13, indicating the first line in accordance with the rules.  But what about those three new changing lines?  Ignore them; they are not allowed to change.  Create the new hexagram based solely on the original result.  Most of the examples are in fact treated this way.  However, there is no mention of any extra changing lines in the text, and having to add them along with another rule that says “ignore them” strains the rules’ credibility.  Richard Rutt says of this objection that “while it shows the Nanjing rules may be fallacious, it does not prove that they are wrong.”  Correct; but it does nothing to convince me, either.

(Note that if the Nanjing rules are strictly applied, our conventional ideas about how to create a second hexagram in the presence of multiple changing lines must be done away with, since in this system most of those lines are ignored.)

My last objection is that there appears to be more than one method of divination represented in the examples, and deriving one set of rules from all of them together strikes me as misguided.  As previously stated, most of the examples are simple cases of zero or one changing line.  One of the latter offers a clue as to the type of divination that was used to generate them.  Example 15 reads:

Kong Chengzi divined with yarrow wands . . . .  He obtained Zhun (Hexagram 3).  Then he said ‘I would have Zhi succeed, and that this be acceptable.’  He received the line of Zhun (Hexagram 3) that changes to make Bi (Hexagram 8) . . . .

The text makes no mention of a coincidence of two divinations resulting in the same hexagram.  It is equally possible that the method used involved first selecting a hexagram, then selecting one changing line.  Yarrow stalks can be used this way, as described by several contemporary authors (Jou, 2000; A. Huang, 2004).  One first obtains two trigrams by twice counting off the stalks by eight, each time using the remainder (0-7) as the result.  Then a line is selected by counting off by six (or seven, allowing for the possibility of zero changing lines).  This type of method, by the way, explains the “extra” stalk that is set aside at the beginning:  it changes the numbers from 0-7 to 1-8.

This observation becomes more significant in light of the curious wording of three of the examples, which refer to a divination method resulting in the number 8.  It is assumed that 8 refers to an unchanging, rather than a changing, line, on which the results were based.  Even more intriguing, two of these three, 10 and 20, are hexagrams that do not fit the simple one-line pattern and require special “rules.”  Can anything be concluded from the results?

(1) In example 10, Mu Jiang “cast the yarrow wands and met with an 8 (an unchangeable broken line) in Gen (Hexagram 52).”  Hexagram 52 has four broken lines, all possible 8’s.  No second hexagram is mentioned as part of the divination.  But the diviner (to whom Mu Jiang was speaking) concluded, without explanation, that “This means that Gen gives Sui (Hexagram 17).”  The base hexagram thus has five changing lines, three of them broken (52:1,3,4,5,6, or 689669).  55 - 44 = 11, so according to the rules, the indicated line is in fact the single unchanging line in the second place, an 8.  The only problem (to us today, at least) is that until the identity of the second hexagram is announced, the “8” in the divination cannot be identified.  Even if it were known that the 8 of the divination is the 8 in the second place, 887867 (changing to Hexagram 53) and 887687 (changing to Hexagram 56) are two of many alternatives that fit the rules equally well.  There is more to the divination method being used here than is revealed in the text.

(2) Example 20 states, again with no explanation, that the Marquis of Jin’s son “received Zhun (Hexagram 3) turning to Yu (Hexagram 16) by an 8.”  The base hexagram is thus 3:1,4,5, or 988698; and 55 - 48 = 7; so according to the rules, the indicated line is the 8 at the top.  But there are two other 8’s to choose from; why the rules favor the top one is unknown.

(3) In example 21, Dongyin says that ‘I have divined with yarrow wands and received an 8 of Tai (Hexagram 11) . . . .’  The statement for this hexagram is used as the oracle; there is no second hexagram mentioned.  I believe that the book has a misprint here; the lines are given as 877888, even though the first is an unbroken line; the numbers are evidently meant to be 977888.  55 - 47 = 8, so using these numbers, the indicated line is the fifth, which is in fact an 8.  According to the rules, there being only one changing line, a second hexagram is not created in this case.  However, 997886 is another valid solution to the “equation,” indicating the same 8 in the fifth place.  But this version changes to Hexagram 52 via three changing lines.  What was the method used to specify the result actually recorded?  It is a “black box,” not revealed in the text.

To round out the atypical results:  in example 22, Viscount Xiang of Shan said that ‘I have heard that Jin divined with yarrow wands and got Qian (Hexagram 1) changing to Pi (Hexagram 12)’ (1:1,2,3).  Again, the method used is a mystery.

To carry the devil’s advocacy one step further, we are assuming that “8” in the Zuo and Guoyu results refers to an unchanging broken line.  Note that no other line number is ever mentioned.  In the absence of evidence to the contrary, “8” may refer to something unknown to us.


V.  The Chico school of thought

Of the sixteen divinations recorded in the text, twelve were made by obtaining and changing either zero or one line of the base hexagram.  The interpretations were made accordingly using either the hexagram statement or the changed line.  No further rules are required.

The other four divinations used a different method, described in three cases as resulting in an “8” which may refer to an unchanging line, and in three cases resulting in multiple changing lines.  The actual method is not revealed in the text, and there is thus no way to generalize it to a system of rules.

The Nanjing rules are interesting, and took a great deal of ingenuity to come up with.  But if they were an investment strategy, I would not risk money on them.  I would rather wait for more historical data about Spring and Autumn period divination methods before drawing any conclusions.





I believe that there is one more misprint in the book:  the equation for example 3 should read 55 - 47 = 8; the oracle is the fifth line of Hexagram 14.





Zhouyi
Rutt, Richard.  Zhouyi:  A New Translation with Commentary of the Book of Changes.  London and New York:  Routledge, 2002.