2.1. The characters nǎo and nǎo

The characters nǎo and nǎo each appear once on the Liye Qin Slips; see (1) and (2) in Table 1.

The two figures in (1) and (2) were converted into the modern characters nǎo and nǎo, which are composed of 匕 bǐ, 止 zhǐ, and 山 shān (the latter is composed of 止 zhǐ and 山 shān). It is generally agreed that these two characters are variants of the characters 匘 nǎo and nǎo.

According to the annotator of the Liye Qin Slips, the character 匘 nǎo should be read as the word “glutinous rice, 稬 nuò” by phonetic loan, as the word 稬 nuò means “rice (or rice plant), 稻 dào” in the Pei state 沛國; therefore, the character 匘 nǎo here might represent the word “rice, 稻 dào”.Footnote ⁷

However, the characters 匘 nǎo and 稬 nuò are not interchangeable by phonetic loan. 匘 nǎo came from *-u (or *-aw) in OC, and 稬 nuò descended from *-or (or *-oj). Therefore, it is unlikely that the character 匘 nǎo represents the word 稬 nuò “glutinous rice” here; see Table 2.Footnote ⁸

Table 2. Reconstruction compared: 匘 nǎo, 稬 nuò, and 稻 dào

While 匘 nǎo and 稻 dào bear the same vowel *-uʔ with the same tone category, they do have different initials *n- and *l-. For this reason, 匘 nǎo in the Liye Qin Slips does not represent the word “rice, 稻 dào” directly.Footnote ⁹

In addition, since the character 稻 dào itself appears in the Liye Qin Slips, it does not seem that 匘 nǎo and 稻 dào represent the same word, as seen in Table 3.

Table 3. 稻 dào in the Liye Qin Slips

It is possible that 稻 dào is a standard form for “rice”, while 匘 nǎo is a substratum form (local languages).

In what follows, we will examine related characters to confirm the reconstructed form of (匘) nǎo.

2.2. Related characters in other excavated documents

In addition to the Liye Qin Slips, related texts with (匘) nǎo appeared in other excavated documents (see Table 4).

Table 4. Related characters

As seen in Table 4, the characters nǎo and nǎo represent the word for “brain, 腦 nǎo” in (3), (4), and (5).Footnote ¹¹ The character in (6) represents the word for 柔 rǒu, meaning “supple, soft, and flexible”. As 柔 rǒu unambiguously came from OC *nuʔ according to rhyme evidence,Footnote ¹² is *nuʔ (or *nˤuʔ) as well. As discussed below (§2.3), the bottom part of the character , which is , is the phonetic element. Note that the character nǎo is considered a variant of (see §2.3 for details). Based on these phonetic loans, the character nǎo and 腦 nǎo simultaneously can be reconstructed as *nˤuʔ in OC.Footnote ¹³ See the reconstructed form in Table 5.

Table 5. The reconstructions compared: 柔, 腦, and ()

2.3. The development of the character (匘) nǎo and its phonetic element

In addition to the Liye Qin Slips, Chen Jian (Reference Chen2008/2019) provided another document that records a character related to nǎo in the Western Zhou Bronze scripts (dated at the latest to the late Western Zhou);Footnote ¹⁵ see Figure 2 and Table 6.

Figure 2. Liang Ji Guan 梁姬罐 (Tomb M2012 of Guo State 虢国墓地 M2012)

Table 6. The character in the late Western Zhou bronze scripts

The character in (7) is composed of the elements 米 mǐ, 艸 cǎo, and 刀 dāo. The original annotator of Liang Ji Guan pointed out that the right-hand element of the character (艸 cǎo and 刀 dāo) is a pictograph meaning the act of “cutting the grass with a knife”.Footnote ¹⁶ Notably, Chen Jian (Reference Chen2008/2019: 97–105) presented a hypothesis that the character in (7) is an antecedent of the character nǎo. The transformation of the shape from to 腦 nǎo is summarized in Table 7.

Table 7. The transformation of the character 腦 nǎo

The middle part of the character in (8) has transformed to in (9) and to in (10), (11), and (12). These characters seem to have been transformed to 止 zhǐ or 山 shān in the Qin and the early Han dynasties. Finally, the character nǎo is transformed to , as shown in (13); see Table 8.

Table 8. The development of the parts of nǎo

In addition, the characters 刀 dāo in (8) and (9) are replaced by 匕 bǐ in (11) and (12); see Table 9.

Table 9. The change from 刀 dāo to 匕 bǐ

Since the characters nǎo and nǎo almost always represent the word for “brain”, the semantic unit “肉 (月)” might have been added instead of “匕” as in (13).

As seen in the contexts in (1) and (2), it is generally agreed that nǎo represents a kind of grain; however, what kind of grain it stands for precisely remains unclear. The shapes of the characters might be key to determining their evolution. The character nǎo was regarded as a pictograph meaning the act of “cutting the grass with a knife”. If true, the character nǎo might originally have represented the word “cutting the rice plant” or the noun “rice plant” and then changed so that it could also represent “husked rice” and “unhusked rice” at the latest shortly before or after the Qin dynasty.

In contrast, the components of the character 稻 dào, as Sagart (Reference Sagart2011: 128) has argued, have the meaning of “dehusked rice grains out of the mortar”.Footnote ²² Compare the characters shown in Table 10.

Table 10. nǎo and 稻 dào

According to the components, the former probably initially meant “rice plant” and the latter “husked rice (or its process)”. The character 稻 dào initially referred to “husked rice” and changed so that it could also represent “rice plant” at a later stage.

Table 11. Semantic changes

In section 3, we examine the word “稻 dào” in OC and the Hmong-Mien languages (hereafter HM).

3.1. 稻 Dào in OC

稻 Dào is reconstructed as *[l]ˤuʔ based on the rhyme evidence in the texts of the Shījīng and Xiéshēng connections.Footnote ²³

Previous studies considered the Chinese word for “rice (or rice plant), 稻 dào” to be a loan word from HM. For example, Haudricourt and Strecker (Reference Haudricourt and Strecker1991: 335–41) expected that agricultural words were borrowed from HM and that “husked rice” was one of them.Footnote ²⁵ Haudricourt and Strecker regard these words as HM substrata in Chinese. Ratliff (Reference Ratliff2018: 131–2) adduced three reasons why rice is regarded as a loan from HM into Chinese: (1) “… the archeological record shows that rice cultivation occurred in the south before it was known to the ancient Chinese,Footnote ²⁶ who cultivated millet, and that the ancestors of the Hmong-Mien people were in the right location to have been the first rice cultivators”;Footnote ²⁷ (2) “… there is no evidence of an initial nasal in Old Chinese”; and (3) “… the word does not appear in Tibeto-Burman”. There is no internal evidence for rice (or rice plant) to reconstruct the prenasalized obstruent in Chinese. Hence, it is highly probable that the Chinese language adopted the word “rice (or rice plant)” from Proto-HM.Footnote ²⁸ However, the tone is irregular; the OC Shǎngshēng (rising tone) tone typically corresponds to Tone B in HM.Footnote ²⁹

Table 12 shows the “rice plant” data for Hmongic languages.Footnote ³⁰ As demonstrated, the proto-form of onsets for “rice plant” is reconstructed as *mbl- based on the data.Footnote ³¹

Table 12. “Rice plant” data in Hmongic, from Ratliff Reference Ratliff2010: 48Footnote ³²

If “rice (or rice plant), 稻 dào” is a loanword from HM, it is assumed that the prenasalized obstruent was dropped and simplified to *[l]ˤ- when the ancient Chinese people borrowed it (*mbl- > *[l]ˤ-).

Notably, the word for “rice plant” in Proto-Min (hereafter P-Min) is reconstructed as *diu^tone2 粙. P-Min Tone 2 corresponds with OC Shǎngshēng, with no problem with tonal correspondences. The problem is that the P-Min form has a semivowel -i-, whereas the MC 稻 dào belongs to division 1, which does not reflect -i-. The semivowel -i- in Min might have come from *dl- within certain stages of P-HM as follows: *mbl- > *ndl- > *dl- > *dj- (P-Min *diu^tone2).Footnote ³³ The semivowel *-i- indicates that P-Min might have borrowed the word “rice plant” from a different stratum with OC and that the word for “rice (rice plant)” was borrowed from HM.

3.2. The distribution of “rice plant” among the Hmongic languages

The data from Table 12 are converted into Figure 3. This map shows minimal data only.Footnote ³⁴

Figure 3. “Rice plant” in Hmongic

In the southern parts (places 4 and 6), prenasalized clusters remain (*mbl- > mpl-), while in East Hmongic and North Hmongic (places 1 and 2), the dental nasal *n- appears more often (*mbl- > n-).Footnote ³⁵ In addition to these data, other dialects have a prenasalized dental obstruent, such as ndl- and ntl-.Footnote ³⁶ These data show that *mbl- had assimilated to ndl- and ntl- before the lateral *-l-.

3.3. Nǎo in the Liye Qin Slips and “rice plant” in North Hmongic languages

In a fascinating detail, no. 0 in Figure 3 indicates the location of Liye Gucheng where the Liye Qin Slips were unearthed. It is obviously near no. 2 “Jiwei 吉衛 (North Hmongic)” with the dental nasal *n- for onsets. The straight-line distance between Liye Gucheng and Jiwei is only 50 km. Most of the data from the North Hmongic subgroups have the nasal n- for unhusked rice (or rice plants), as shown in Table 13.

Table 13. Data from North Hmongic: unhusked riceFootnote ³⁷

Figure 4 shows the distribution of the data from North Hmongic (no. 0 indicates the location of Liye Gucheng).

Figure 4. Rice in North Hmongic

As mentioned in §2 above, the characters nǎo and nǎo are reconstructed as *nˤuʔ, and this intriguingly coincides with Jiwei [nɯ²], Yangmeng [nɯ³¹], and Zhongxin [nɯ³⁵]. Hence, it is assumed that the character nǎo in the Liye Qin Slips represents the same word as “rice or rice plant” in these languages. Although it is difficult to determine whose word it was originally, if this assumption holds, we can deduce that OC borrowed at least two varieties of “rice plant”: *mbləu in P-HM (or P-Hmongic) and *nəu in “Proto-North Hmongic (provisional) or certain groups of North and East Hmongic speakers”.Footnote ³⁸ The former dropped *mb- when the Chinese borrowed it; the latter is a substratum word, and the word itself was lost in the ancestor of Middle Chinese and modern Chinese dialects (when the character nǎo became used for the word “brain 腦 nǎo” instead), as shown in Table 14.

Table 14. The direction of loanwords: OC

As mentioned above, the P-Min “rice plant” is reconstructed as *diu^tone2. This could have come from the different HM dialects with OC, as shown in Table 15.

Table 15. The direction of loanwords: P-Min “rice plant”

Since the characters nǎo and nǎo, which initially referred to “rice or rice plant”, represent the word for “brain, 腦 nǎo” bearing the onset *nˤ- in the Qin Slips and the Mawangdui silk manuscript, we can estimate that the assimilation (*mbl- > *n-) must have taken place at the latest sometime before or after the Qin dynasty (approximately 200 K) in East and North Hmongic. Proto-Hmong-Mien might be even older.Footnote ³⁹