In these posts, we are looking at an itinerary through Southeast Asia that Chinese scholar-official Jia Dan recorded in the ninth century. Why is this important?
It is important because although there is considerable information in Chinese sources about early Southeast Asia, it can be difficult to determine from that information where exactly certain places were. That is why Jia Dan’s itinerary is so important, as it shows places in relation to each other.
Therefore, if we follow Jia Dan’s itinerary, and check the placenames that he mentions against other mentions of those same placenames in other sources, we can develop an understanding of which places certain Chinese placenames actually referred to.
In examining how Paul Pelliot (1904) and Geoff Wade (2013) have written about this itinerary, we see that both of these scholars pictured a route that went directly down to the Strait of Malacca from the end of the Indochinese Peninsula, and then over to the island of Java.
After reaching that point, Pelliot, Wade, and every other scholar who has ever followed the itinerary into the Strait of Malacca get lost.
As I pointed out last time, Jia Dan recorded a route from Guangzhou to India, and no matter how one imagines such a route, it makes absolutely no sense to sail to the island of Java. However, that is what Pelliot argued, but then later in his article, when he discussed the next leg of the journey, he backtracked. Let us look first at the next leg of the journey, and then we will examine what Pelliot wrote.
Onward to Gegesengqi
Again, the place that we left off discussing was a place called Heling, which Pelliot strongly argued was “Java.” The itinerary continues as follows:
又西出硤，三日至葛葛僧祇國，在佛逝西北隅之別島，國人多鈔暴，乘舶者畏憚之。其北岸則箇羅國。箇羅西則哥谷羅國。
Puis, vers l’ouest, sortant du détroit, après trois jours on arrive au royaume de Ko-ko-seng-tche [Gegesengqi], qui se trouve sur une île séparée à l’angle nord-ouest du Fo-che [Foshi]. Les hommes de ce royaume sont pillards et cruels; les navigateurs les craignent. Sur la côte septentrionale, c’est le royaume de Ko-lo. A l’ouest du Ko-lo, c’est le royaume de Ko-kou-lo. (373)
Then, westward, on leaving the strait, after three days one arrives at the kingdom of Gegesengqi 葛葛僧祇, which is located on a separate island at the northwestern corner of Foshi 佛逝. The people of this kingdom are pillagers and cruel; navigators fear them.
On its northern shore is the kingdom of Guluo 箇羅. West of Ko-lo is the kingdom of Geguluo 哥谷羅.
Before discussing this passage, Pelliot “reorients” his reader by stating the following:
Jia Dan’s itinerary did not pass through Java. It was because, before crossing the Straits, many travelers made a detour to that island that he mentioned it as lying to the east of Fo-che [Foshi], that is, Sumatra. But it is to the strait that had Lo-yue [Luoyue] on its northern shore and Fo-che [Foshi] on its southern shore that the itinerary returns in order to give the continuation of the stages between China and Ceylon. (348)
Whoa, whoa, whoa! Slow down! I think we need to go back and look again at Pelliot’s translation of the itinerary up to this point.
又五日行至海硤，蕃人謂之「質」，南北百里，北岸則羅越國，南岸則佛逝國。佛逝國東水行四五日，至訶陵國，南中洲之最大者。又西出硤，三日至葛葛僧祇國，在佛逝西北隅之別島. . .
Then, after five days’ travel [from the end of the Indochinese Peninsula], one arrives at A STRAIT [un détroit] which the barbarians call Zhi 質. From north to south it is one hundred li. On the northern shore is the kingdom of Luoyue 羅越; on the southern shore is the kingdom of Foshi 佛逝. To the east of the kingdom of Foshi, traveling by water for four or five days, one reaches the kingdom of Heling 訶陵; it is the largest of the islands of the south. Then, westward, on leaving THE STRAIT [du (= de le) détroit], after three days one arrives at the kingdom of Gegesengqi 葛葛僧祇, which is located on a separate island at the northwestern corner of Foshi. . . (348-49)
What Pelliot is arguing/reading here is that one traveled down to somewhere at the end of the Malay Peninsula where there was “a strait” (un détroit) and then went to Foshi, which he argues was on the eastern coast of the island of Sumatra.
After that, he now states here that the information about the journey to Heling/Java was just “FYI.” The real journey continued on from “the strait” (du [= de le)] détroit), meaning the strait that one reached before proceeding on to Foshi.
Ok, but, 1) in reaching Foshi, one would have already “left the strait,” so it doesn’t make sense to then go back to the strait to leave it again and continue the journey.
2) The distinction between “a” strait and “the” strait is not evident in the Chinese text.
What this shows is that Pelliot, like everyone who followed him in interpreting this itinerary as heading from the Indochinese Peninsula to the Strait of Malacca and Java, was starting to struggle here to get this itinerary to fit the geography of that area.
Let’s now look at how he explained the next part of the journey.
“Then, westward, on leaving the strait,” says Jia Dan, “after three days one arrives at the kingdom of Gegesengqi 葛葛僧祇, which is located on a separate island at the northwestern corner of Foshi 佛逝. The people of this kingdom are pillagers and cruel; navigators fear them.”
. . .
This part of the text is disconcertingly obscure and, as it stands, seems difficult to justify. The explanation I am about to propose is only a makeshift.
The first question that arises is where Gegesengqi was located. If one admits that Jia Dan took Foshi as a name for the whole island of Sumatra, then Gegesengqi must be sought on one of the small islands, Pulo Bras or Pulo Way, which lie off the northwestern tip of the main island. However, particularly in the Straits, where the monsoon blows only weakly, it seems impossible that junks could go in three days from the entrance to the Straits to Pulo Way. Moreover, it would then become very difficult to place to the west of Pulo Way or Pulo Bras the various countries that Jia Dan still enumerates before arrival at Ceylon. I therefore prefer to take Foshi in the second sense that this name has in the History of the Tang: it is the name of the whole island of Sumatra, but it is also, more specifically, the name of the southeastern part, while the northwestern part is called Poluoshi. On this interpretation, Gegesengqi would be one of the Brouwers Islands; I have no information that would allow me to be more precise. The name Gegesengqi remains mysterious to me; perhaps the final part should be read sengqi; but this term, which properly designates negroes, is of no help to us here. (348-349)
As I said, everyone who tries to get this itinerary to go through the Strait of Malacca gets lost at this point.
The term, Foshi, does not have a “second sense” (seconde valeur). Instead, what Pelliot states about Foshi comes entirely from his imagination, not from any Chinese source.
Indeed, with a statement like “If one admits that Jia Dan took Foshi as a name for the whole island of Sumatra,” we enter the world of fantasy, as we can clearly see from all of the early sources that Chinese did not have a conception of such a vast area of geographic space.
To take an example, let us look at what the Tang-dynasty monk Yijing 義淨 recorded about the region in the late seventh century, having traveled through it himself, just a century before Jia Dan recorded the information about the itinerary.
然南海諸洲有十餘國，純唯根本有部，正量時欽，近日已來少兼餘二(從西數之，有婆魯師洲、末羅遊州，即今尸利佛逝國是。莫訶信洲、訶陵洲、呾呾洲、盆盆洲、婆里洲、掘倫洲、佛逝補羅洲、阿善洲、末迦漫洲，又有小洲不能具錄)，斯乃咸遵佛法，多是小乘；唯末羅遊少有大乘耳。諸國周圍，或可百里、或數百里、或可百驛；大海雖難計里，商舶串者准知，良為掘倫。
However, among the various islands of the Southern Sea there are more than ten kingdoms. They follow exclusively the Mūlasarvāstivāda, while the Saṃmatīya are also respected at present; in recent times a few also combine the other two [traditions]. (Counting from the west, there are Polushi Island, Moluoyou Island—that is, what is now the kingdom of Shilifoshi—Mohexin Island, Heling Island, Dandan Island, Panpan Island, Poli Island, Juelun Island, Foshi-buluo Island, Ashan Island, and Mojiaman Island; there are also smaller islands that cannot all be recorded.) Thus, they all revere the Buddha’s Dharma, and most are followers of the Lesser Vehicle; only in Moluoyou is there a slight presence of the Great Vehicle.
As we can see here, EVERY PLACE Yijing visited was an “island” (zhou 洲) to him.
Is that because all of these places were actual islands? No, it is because he did not have a birds-eye or aerial view of the region, and only saw each place from the water level, and at that level, it was impossible to discern the actual geographic shape of the region.
As such, not only was Foshi not conceived as something as large as the entire island of Sumatra, but as we will see below, it was also not conceived as “the name of the southeastern part” of that island either.
What led Pelliot to imagine what he did is the reference to “Foshi,” and then later to Gegesengqi as “a separate island at the northwestern corner of Foshi.”
The key information that has to be deciphered is this: From Foshi you go east to Heling, and then in the west you exit a strait and go to Gegesengqi which is in a northwest corner of Foshi.
This puzzle cannot be deciphered by looking at places in the Strait of Malacca. It only makes sense by looking at the Malay Peninsula.
THE Southeast Asian trade Corridor
For centuries prior to the time that Jia Dan recorded his itinerary, THE one and only trade corridor in the region extended from Cambodia to the area of what is now Surat Thani province in Thailand, from where trans-peninsular trade routes went across the land at the Isthmus of Kra and from Surat Thani to what is now Phang Nga province where the archaeological site of Takua Pa is located.
As we saw in Part 1 of this series, that trade corridor is referenced through the mentioning of the main trade centers on either end of that trade corridor, such as Funan 扶南 [in Cambodia], and Dunxun 頓遜 (or Dianxun 典遜) [in Surat Thani province], or Luoyue 羅越 [in Cambodia] and Geguluo 哥谷羅 [in the Takua Pa area].
In Jia Dan’s itinerary, Foshi was also in that area of Surat Thani province, and Gegesengqi, as “a separate island at the northwestern corner of Foshi,” was on the western side of the Malay Peninsula at the terminus of an overland trade route from Foshi.
However, to be able to see this, we have to first cure ourselves of a certain malady that scholars have long suffered from concerning their understanding of “maritime” travel during the time period we are examining.
Understanding Tang-era Travel
The idea that people traded and traveled across the Malay Peninsula has been difficult for scholars to accept. At the core of the problem, I would argue, is a misunderstanding of what early Chinese texts record about travel.
In particular, if scholars think that someone was traveling by sea, then they generally assume that every part of that journey was on the sea. However, there is no reason to think that, and we can understand that if we consider what is not recorded about the counterpart of “maritime” travel, that is, “overland” travel.
So, in the case of Jia Dan’s itinerary, because it departs by ship from Guangzhou, scholars assume that either 1) that same ship would go all of the way to India (highly unlikely), or if they consider that travel at that early time was done by taking different ships for different legs of a long journey, then 2) they still assume that every leg of the journey was on the water in a ship.
Meanwhile, the same scholars never seem to consider or question how one traveled on the overland routes that Jia Dan documented. For instance, with regards to the route from central Vietnam to Cambodia, how did people travel across that geographic space? Since we see it as “overland,” are we to assume that they walked on the “land” the entire way? Did they not sometimes get in a boat and go down a river? That would have been a much more efficient way to travel “overland” than walking. Did they perhaps sometimes ride an elephant? Or get carried in a sedan chair?
We have no idea because Jia Dan, like so many other chroniclers, did not record that granular-level information. He just told you the order of places that one passed by.
Jia Dan, like other chroniclers from this period, recorded an itinerary. He did not write a travel guide.
To take another example, there is a record in the History of the Liang (635) of a place on the northern part of the Malay Peninsula, probably in the area of the Isthmus of Kra, called Dunxun that states the following:
頓遜之東界通交州，其西界接天竺、安息徼外諸國，往還交易。所以然者，頓遜迴入海中千餘里，漲海無涯岸，舶未曾得徑過也。其市東西交會，日有萬餘人。珍物寶貨，無所不有。
On its eastern side Dunxun communicated with Jiaozhou [the area of northern Vietnam]; on its western side it connected with Tianzhu [“India”], Anxi [“Parthia/Iran”], and countries beyond the frontier, with traders going back and forth in commerce.
The reason for this was that Dunxun jutted deep into the sea for more than a thousand li, while the Swelling Sea [Zhanghai 漲海] had no shores, and ships had never been able to pass straight through.
Its market was where east and west met; each day more than ten thousand people were there. There was no precious good that it did not have.
First, note the use of the term “Swelling Sea.” As we saw in the first post, this was a term that was used prior to the Tang dynasty period to refer to the Gulf of Thailand area, and notice what a formidable and unknown body of water it was. It was too vast for Chinese mariners to sail across.
Second, how exactly did this interaction between Dunxun, on the eastern side of the Malay Peninsula, and places to the west of the peninsula take place?
Well, since the only way to get from “India” to the Malay Peninsula would have been by first departing on a ship, which means that one was on a “maritime” journey, and since there is no information here about anyone “walking,” does that mean that we can only conclude that people sailed from India through the Strait of Malacca and then up the east coast of the Malay Peninsula?
NO!! Of course not. We know that people sailed along the coasts in these early centuries (and the statement about the Swelling Sea confirms that), and we know that there was an overland passage across the Malay Peninsula at the Isthmus of Kra, and therefore, it is completely logical to assume that the contact between Dunxun and places to its west took place through overland travel and trade.
Yet, for some inexplicable reason to me, this common-sense fact has been extremely difficult for scholars to grasp for well over a century now. While they might be able to accept that the above passage indicates the presence of trans-peninsular trade, we are supposed to believe that less than 60 years after this text was compiled, the monk Yijing sailed through the Strait of Malacca (I will discuss his itinerary in later posts), and that a century still later, Jia Dan recorded an itinerary that went directly south from the Indochinese Peninsula to the Strait of Malacca. . .
Do we have evidence of a major improvement in nautical technology in that time period? No.
The New Trans-Peninsular Crossings in the Tang Period
Again, while some scholars can see evidence of a trans-peninsular crossing getting used in the area of the Isthmus of Kra prior to the Tang dynasty period, they then interpret Tang-era sources as indicating travel through the Strait of Malacca.
That, of course, would point to a massive advance in nautical knowledge and technology in a very short period of time, but as we saw in Part 1, Pelliot’s attempt to get Jia Dan’s itinerary to extend from the Indochinese Peninsula to the Strait of Malacca was deeply flawed.
By contrast, I argued that Jia Dan’s route followed the southern coast of the Indochinese Peninsula (i.e., still avoiding the “Swelling Sea”), and then crossed over to the area of Surat Thani at the western end of the Gulf of Thailand.
In doing so, the itinerary up to that point followed a well-known route, as this was the area of the trade corridor that had been employed for centuries. And Foshi was in that area of Surat Thani province at one end of a trans-peninsular trade route, with probably Takua Pa at the other end.
The emergence of “Heling” in Tang period sources, however, marks something new. It was a place that was reached in five days by heading east from Foshi. If one follows the coast from Surat Thani, one first heads directly to the east, and then the coast heads to the southeast.
Five days down that coast from Surat Thani would have put one somewhere in the greater Lake Songkla region, a region that was more easily accessible at that time than it is today, as the area between what is now Sating Phra and the Malay Peninsula had not been closed in by silting.
Further, the greater Lake Songkhla region was home to multiple trans-peninsular trade routes, some of which, such as the one from Lake Songkha to Kedah, and probably the one from what is now Phatthalung to Trang, were perhaps easier to cross than the route from Surat Thani province to Takua Pa.
This area became the center of trans-peninsular trade in the Tang dynasty period, and this is an incredibly logical development that EVERY SCHOLAR has missed.
It is not the case that there was a sudden change over the course of a century or two from trading across the Isthmus of Kra to sailing through the Strait of Malacca. Instead, there was a gradual development over the course of a millennium from trading across the peninsula at the Isthmus of Kra, to trading across the peninsula at the greater Lake Songkhla area, to sailing through the Strait of Malacca.
My Reading of Jia Dan’s Itinerary
Let us now return to the itinerary and I will explain how I understand it.
又五日行至海硤，蕃人謂之「質」，南北百里，北岸則羅越國，南岸則佛逝國。佛逝國東水行四五日，至訶陵國，南中洲之最大者。又西出硤，三日至葛葛僧祇國，在佛逝西北隅之別島. . .
Then, after five days’ travel [from the end of the Indochinese Peninsula], one arrives at a “sea strait” [haixia 海硤] which the barbarians call Zhi 質. From north to south it is one hundred li. On the northern shore is the kingdom of Luoyue 羅越; on the southern shore is the kingdom of Foshi 佛逝. To the east of the kingdom of Foshi, traveling by water for four or five days, one reaches the kingdom of Heling 訶陵; it is the largest of the islands of the south. Then, westward [or “in/to the west”], on leaving “a strait” [xia 硤], after three days one arrives at the kingdom of Gegesengqi 葛葛僧祇, which is located on a separate island at the northwestern corner of Foshi. . . (348-49)
Unlike Pelliot, I see no reason to take an FYI detour to Heling. This is all talking about a continuous route. As such, there is no need to indicate “a” strait and “the” strait, because this passage refers to two different straits.
That said, there is a difference in how these two straits are recorded. The first is record as a “sea strait” (haixia 海硤) and the second is just recorded as a “strait” (xia 硤).
This term, xia 硤, is somewhat uncommon, and is defined as “a stream valley between two mountains” (兩山之間的溪谷), and its more common variant, xia 峽, is glossed in English as “gorge, strait, ravine; isthmus.”
In an earlier working paper, I tried to point out that in the first instance that this term appears in the above passage, it is modified by the character hai 海, meaning “sea,” whereas in the second instance it is not.
At that time, for the sake of simplicity, I chose to translate these two terms as “sea pass” and “mountain pass” rather than “a sea stream valley between two mountains” and “a stream valley between two mountains,” to try to emphasize that the basic meaning of this term refers to a land context, that is, to “a stream valley between two mountains,” and I thought that this second instance could be a reference to the trans-peninsular route from Phatthalung westward which indeed passes through “a stream valley between two mountains.”
Johannes Kurz objected to this wording in some comments that he posted on Academia.edu, and argued instead that xia means “gorge” and then went on to argue that it was a reference to another strait, although he provided no information about where that strait might have been.
In fact, if we look at the basic meaning in Chinese, “a stream valley between two mountains,” we don’t necessarily get “gorge,” but gorge could also work for the area to the west of Phatthalung. However, having said all of this, it doesn’t matter if this second reference is to “a stream valley between two mountains,” “a gorge,” or “a strait,” as I have translated it this time, as any one of these translations will work for the route that I see this itinerary following.
In particular, when we realize that Jia Dan was, like everyone before him, describing a route that followed the trade corridor from Cambodia to Surat Thani, and that he then headed into new territory slightly to the south of that area, then we can solve the puzzle of “From Foshi you go east to Heling, and then in the west you exit a strait and go to Gegesengqi which is in a northwest corner of Foshi.”
Jia Dan was documenting the following route.
In terms of movement, Jia Dan clearly mentions “traveling by water for four or five days, one reaches the kingdom of Heling,” and “on leaving a strait, after three days one arrives at the kingdom of Gegesengqi.
What he doesn’t mention, is how one got from Heling to “the west” or “in the west” or “westward.” This is not surprising, as the act of crossing the Malay Peninsula was rarely mentioned in Chinese sources, and the passage from the History of the Liang is an example of this.
Finally, Gegesengqi’s position as “a separate island at the northwestern corner of Foshi” makes complete sense when we understand that Jia Dan was describing the itinerary depicted in the image above.
Gegesengqi was at the western end of a trans-peninsular trade route that crossed from Surat Thani, and Foshi was on the eastern end.
If that was the case, then why, you might ask, didn’t Jia Dan just record a route directly from Foshi to Gegesengqi?
I think the answer to that can be found in what I said above: the trans-peninsular routes from Lake Songkha to Kedah, and probably the one from what is now Phatthalung to Trang, were perhaps easier to cross than the route from Surat Thani province to Takua Pa.
As we will see later, the monk Yijing traversed the Surat Thani trans-peninsular route. A century later, Jia Dan recorded one of the routes across the peninsula from the Lake Songkhla region. In addition to documenting an itinerary, Jia Dan was actually also marking a change in power in the region. He was documenting the rise of the Lake Songkhla region to prominence, something that the Arabic accounts of the ships that stopped at Kedah did as well.
Geoff Wade’s 2013 Translation
Finally, as we have done in the previous two posts, let us look at how historian Geoff Wade translated the passage here to gain an understanding of how Pelliot’s ideas changed, or didn’t change, over the course of the twentieth century.
Further west one exits the strait. And within three days one reaches the country of Gegesengzhi. 26 This is another island to the north-west of Foshi. The inhabitants of this place engage in plunder and those on passing ships greatly fear and dread them. (88-89)
26. Unidentified polity.
As we saw in Part 2, Wade had the itinerary go west from Foshi to Heling, which he argued was Java, when the text says that one went to the east from Foshi to reach Heling. Further, it is unclear what he thought “the strait” here referred to, as unlike Pelliot, he did not offer an explanation.
More significantly, however, his translation about the location of Gegesenqi that “This is another island to the north-west of Foshi” is not accurate. The text records that Gegesengqi was “on a separate island at the northwestern corner of Foshi” (在佛逝西北隅之別島).
In this itinerary, Foshi is first reached from the east, and then when Gegesengqi is later mentioned, it is located in the “northwest corner” of Foshi. Hence, what is recorded here is an awareness of the expanse of Foshi from east to west, one that was attained by crossing the peninsula and proceeding to a place that was on the western end of a trade route that went across the area under the authority of Foshi, based in the area of Surat Thani.
In imagining that Jia Dan’s itinerary went though the Strait of Malacca, and that Foshi was somewhere near Palembang, it is impossible to see how an understanding of Foshi’s east-west expanse could be attained without doing what Pelliot and Wade both did, which is to not follow what the text actually records, but to just imagine that Gegesengqi was “up there” somewhere.
As such, by fudging this translation, Wade avoided the problem of having to explain the actual spatial relationship that is recorded in the text between Foshi and Gegesengqi.
Once again, we see that over a century after Pelliot published his article in 1904, there was no scholarly advancement.
Absolutely ZERO improvement.
You have nailed it. Agree on the location of Foshi, Heling, could it be Talung, then with open access to the gulf, and the second strait, could it be the strait of Tantalam? There definitely was a strait, but doesn’t really fit in with the itinerary, if Gegesengqi was reached in five days?
Unrelated perhaps, the last two characters of Gegesengqi,僧祇 are the same characters used to describe the residence of the king of 赤土,as in 僧祇城. Are there any other instances where this combination of characters was used as a place name? It is also part of 摩訶僧祇律 that 法顕 translated before his death.
Thanks for the comment!!
It doesn’t surprise me that you can see what I am saying, as you obviously know the sources and you have looked “across” sources to try to trace the meaning of terms, etc. If you look at what I just wrote in Part 4, that is exactly what has not been the case for over a century. Instead, scholar after scholar has just worked from the (flawed) translations in secondary scholarship.
In my mind, I have imagined Heling to be Talung, but I don’t think we have evidence of Talung until many centuries later, and when we do, it’s from Thai, as far as I know. So, I think it’s hard to say. It’s not impossible, but the case would be a lot stronger if there was evidence from in between.
In Arabic sources from roughly the same time as Jia Dan’s itinerary, they often mention three places together, Zabag, Harang and Serendip. Reconstructions of Chinese all seem to have Heling still sounding something like Heling 1,000 years ago, however in Vietnamese, for which it’s Sinitic terms are mostly supposed to date from the Tang dynasty period, it would be Hà Lăng. I don’t know if that “lăng” pronunciation for “ling” is old or a later development, but if it is old, then Harang would certainly appear to come from something sounding like “Halang.”
The “sengqi” question is complicated. There is an idea that it comes from Arabic/Persian and that it means Zanj, referring to a place in East Africa, and that it somehow was used to refer to black slaves that were brought from there. That is a modern explanation, and as far as I know, there is no such explanation in premodern Chinese sources. Instead, you have the monk Huilin 慧琳 say in his 807 dictionary, the Yiqiejing yinyi 一切經音義, the following:
崑崙語＝時俗語便亦曰骨論。南海洲島中夷人也。甚黑裸形，能馴伏猛獸犀象等。種類數般，即有僧祇、突彌、骨堂、閤蔑等，皆鄙賤人也。國無禮義，抄劫為活，愛啖食人，如羅剎惡鬼之類也。言語不正，異於諸蕃。善入水，竟日不死。
“The Kunlun language”—in current vulgar speech this is also called Gulun. It refers to the barbarian people living among the islands and islets of the Southern Seas. They are very black and go about naked. They can tame and subdue fierce beasts such as rhinoceroses and elephants. There are several kinds among them, including the Sengqi, Tumi, Gutang, and Hemie. All are base and lowly people. Their countries have no ritual or righteousness; they live by plunder and robbery, and delight in chewing up and eating human beings, like rakṣasas or evil demons. Their speech is not correct, being different from that of the various foreign peoples. They are skilled at entering the water, and can stay there all day without dying.
I think “the Sengqi” were probably on the western side of the northern part of the Malay Peninsula. The most “uncivilized” people mentioned in Chinese sources were in the area of the Andaman islands. It could be people from there who were captured and delivered to the Chinese court as slaves, and that reference to Chitu (which I’ve long felt is Kedah) could be because it was in that general area. The same passage also refers to it’s citadel as “Lion citadel.” I’ve wondered if it was ruled over by people from the “Lion kingdom” (Sri Lanka), and that it was also called the Sengqi citadel because the people were dark-skinned like the Sengqi people in that area.
That’s a guess, but there is other (indirect) evidence to place Gegesengqi up near Phang Nga/Takua Pa, which I’ll write about later.
The pronunciation of “Chitu” during that period could have been something like “Tshaijk-thuə” (I can’t quite replicate this with all of the specialized linguistic characters that Pulleyblank uses). I think that is “Kedah” (which Yijing writes as Jiecha 羯茶; Pulleyblank = Kiat-trha) and the sound is slightly off because someone was also trying to capture meaning as well, “red earth.” The land near the Bujang Valley has lateritic “red earth” which one would see while traveling overland to get there.
Finally, as for sengqi being a part of 摩訶僧祇律 that 法顕 translated before his death, this is what the Digital Dictionary of Buddhism has (http://www.buddhism-dict.net/ddb/ when it asks for a long-in, just sign in as “guest.” You can search as a guest for like 20 items a day):
僧祇
Basic Meaning: incalculable
Senses:
An abbreviation of 阿僧祇 (Skt. asaṃkhya). [Charles Muller]
An abbreviation for Mahāsāṃghika 摩訶僧祇律. [Charles Muller]
Sanskrit sāṃghika, relating to a saṃgha; a complete set of land and buildings for a monastery. [Charles Muller; source(s): Soothill]
I don’t think there is any connection between the Buddhist term, sengqi, and the usage of those same two characters in the Southeast Asia context.
This relates to an important point, however, and that is that since the time of the first Western examinations of these sources, there has been a tendency to see something Indic in these early names. However, some clearly show something “Southeast Asian.” So, for instance, anything “puluo” is always interpreted as the Indic “pura” and never as the Austronesian “pulau” (island). And the “ge” or “gu” that are at the beginning of many names, like Gegesengqi, are always left untranslated, and people have never considered the very common Mon-Khmer for an island, “ko.”