The earliest record that tells us something about life in the Red River Delta in ancient times is Li Daoyuan’s sixty-century Shuijing zhu 水經注 (Annotated Classic of Waterways). That book cites an earlier work, the late-third or early-fourth-century Jiaozhou waiyu ji 交州外域記 (Annotated Classic of Waterways), to say the following about agricultural practices:

“In the past, before Jiaozhi had commanderies and districts, the land had lạc fields. These fields followed the rising and falling of the. . [KEY WORD]”

交趾昔未有郡縣之時,土地有雒田,其田從潮水上下. . .


The next word in that sentence is very important, and it has caused a lot of confusion. It is chao/triều 潮. The basic meaning of this word is “tide,” and scholars have understood this sentence as saying something about tidal waters.

In 1918, for instance, French scholar Henri Maspero, translated this passage as follows:

“In the past, before Jiaozhi had been divided into commanderies and districts, its territory formed lạc fields, where the water rose and fell following the tide.”

[“Autrefois, au temps où le Kiao-tche n’était pas encore divisé en commanderies et sous-préfectures, son territoire formait les champs lo (lạc) 雒田, où l’eau montait et descendait suivant la marée.” (8)]


Later, in the 1930s, French geographer Pierre Gourou made an extensive, and influential, study of the Red River Delta that he published in 1936 as Les Paysans du Delta tonkinois: Etude de géographie humaine.

In that book, Gourou included a very brief section on tides and their limited ability to spread inland. In that section he stated that,

“The tides may occasion intrusion of salt water into the rice fields, and that is the main danger of them. To be sure there are kinds of rice which can tolerate a rather heavy salt content in their water, but an excessive concentration would ruin the crop. Therefore, rigid diking is necessary to check the salt water.” (pg. 82 of English version)


Then later in the century, we can see the ideas of Maspero and Gourou get repeated in English-language writings.

In 1979 Australian historian Jennifer Holmgren followed Maspero’s wording and wrote in English that “Before Jiaozhou had been divided into commanderies and prefectures, this area formed the fields of the Lo. Here, agricultural labor followed the rise and fall of the tides.” (37)

Then in 1983, American historian Keith Taylor included in his The Birth of Vietnam a map on “Tidal Influence in the Hong River Plain in Modern Times” based on a similar map in Gourou’s book (11), and stated that “The practice of tidal irrigation, as described in the texts that mention Lạc fields, reveals a relatively advanced agricultural technology.” (12)

At the same time, however, in a footnote to that statement, Taylor stated that “Yumio Sakurai has recently conjectured that the description of Lạc fields as tidal is simply a contrived elaboration to explain the name Lạc and that ancient agriculture in Vietnam was based on agronomic expertise, not water-control engineering; this review deserves more study.” (12)


Japanese historian Sakurai Yumio published an article in 1979 in which he argued that early Vietnamese had not made use of the tides in agriculture, that is, that they had not engaged in what can be called “tidal irrigation.”

A summary of his argument can be found in the above image. Essentially he argued that there was no evidence that anything Gourou discussed had existed in ancient times, that the location of the lạc fields was not close to the coast, and finally that different types of rice were grown in the Red River Delta, and that perhaps a variety that was somewhat resistant to salt water had been noticed by Chinese in the past, and that this is what was recorded.

Morohashi 1

While Sakurai’s evidence is important, the character (chao/triều 潮) that scholars have understood to mean “tide” does not have to be understood that way as it had more than one meaning, because it was used as a variant for other characters.

First of all, this character was a variant for chao/triều 𣶃, which was defined in the early-second-century-CE, dictionary, the Shuowen 説文, as meaning “water rightfully retuning to the sea” (水朝宗于海). This definition has a moral sense to it, and the wording of this definition was itself used to symbolize the attitude that vassal lords should have toward the emperor, someone they should always “rightfully return to” just as water “rightfully flowed toward the sea.”

The character, chao/triều 𣶃, was also a variant for tao/đào 濤 or its own variant 𤁟, which the Shuowen defines as “a large wave” (大波也). This is the version that is important for our understanding of that passage.

Morohashi 2

Tidal irrigation is complex and difficult. It requires that one collect salty sea water (that is pushed inland with the tide) in a pool and then add fresh water to dilute the water so that the salt will not kill plants.

What is more, if “agricultural labor followed the rise and fall of the tides,” as Holmgren translated that line in the Jiaozhou waiyu ji, this would mean that people would have had to engage in this complex labor every single day.

However, archaeological evidence of such pools, or of early sea dikes, does not exist.

There is also no solid linguistic evidence either. Indeed, Mon-Khmer languages have very few words for “tide.” This is not surprising, as the Mon-Khmer languages emerged in the inland area of mainland Southeast Asia. Mon-Khmer speakers were not coastal peoples.

mon khmer

As such, rather than seeing this passage as saying that lạc fields were cultivated following the rising and falling of the (daily) tides (chaoshui shangxia/triều thủy thượng hạ 潮水上下), it would make more sense to employ another meaning of chao/triều 潮 and to understand that the lạc fields were cultivated following the (annual) rise and fall of the “wave of water” (taoshui shangxia/đào thủy thượng hạ 濤水上下) that came down the Red River and flooded the fields.

If that was the case, and given the lack of evidence of large irrigation works from this early period, it was probably also the case that the people at that time grew rice by “broadcast seeding” rather than by growing seedlings first and then transplanting them.

So when did full-scale wet-rice agriculture start to take place? I’m not sure, but certainly increasing contact with Tai-speaking peoples in the late first millennium CE, as Southwestern Tai-speakers started to migrate away from the Guangxi region, would have exposed Vietnamese to their complex water control techniques.

And in the period between 1,000 and 1,500 we do find evidence in the Vietnamese annals of major dikes getting built along the Red River.

Perhaps that is when “relatively advanced agricultural technology” really started to be developed in the Red River Delta.