In my report on this in Underworld I address the important issue raised (with reference to Steve's more recent discoveries) by Professor Tsao Nu-ching of Taiwan's Central Geological Survey -- namely that "flood basalts and other types of volcanic lava eruptions can also create linear formations, due to the inherent joints and planar structures of the igneous rocks." (TaiwanNews.com, 26 November 2002)
In the hope that it will help to illuminate this debate, and shed light on the background to the most recent Peng-Hu discoveries here's a bit of what I have to say about it in Underworld (Chapter 30, pages 672-673):
"...the Pescadores [Peng-Hu archipelago] -- located on the tip of a strategic peninsula of mainland China 13,500 years ago, then later one island, then later still the 64 tiny remnants that are seen today -- are a plausible location in which to search for underwater ruins from the flood epoch.
They are plausible for another reason too. Ancient myths of the Pescadores speak of a great castle with huge “red” walls that lies submerged somewhere amongst the islands. It was precisely these myths that led a government official to ask the brilliant Taiwanese diver Steve Shieh to look for underwater ruins if he happened to be working in the area. Over a period of several years, Steve complied, searching the waters around most of the islands. Eventually he was rewarded with an extraordinary discovery off the island of Hu-Ching (“Tiger Well”). This happened more than 20 years ago and has received no attention or publicity in the West. Luckily for me, however, TBS, a large Japanese TV station, ran a report on Steve and his discovery as recently as January 2001. The report was seen by several Japanese friends who drew it to my attention.
We did two days of diving with Steve Shieh off Hu-Ching Island at the end of August 2001.
The structure that he showed us consists of two immense walls, hundreds of metres in length, one running due north-south and the other running due east-west, crossing the north-south wall at right angles. At the north end of the north-south wall is a large circular enclosure, part of which has completely collapsed. The north-south wall is in relatively shallow water -- 4 to 6 metres depth. The east west-wall starts at 4 metres depth but can be followed down to 36 metres depth. All the walls are a consistent height of 3 metres from the base to the top of the wall; however some sections are broken.
In a volcanic, earthquake-prone area such as Taiwan one must be conscious of the possibility that such walls could be natural features -- specifically basaltic dykes (quite common around the Pescadores). Such dykes form when a wall-like mass of igneous rock intrudes into cracks in older sedimentary rock.
Despite extremely strong currents flowing unpredictably from eight different directions, (why are there always currents around underwater monuments?), I was able to examine the walls quite thoroughly. My initial impression is that they are not basaltic dykes. This is mainly so because, after scraping off marine-growth from several sections of the walls Steve showed me courses of individual blocks laid tightly together side-by-side. The joints between the blocks in some cases admit the point of a knife and it was possible for me to work the knife blade in as far as the hilt and move it entirely around individual blocks. In addition the nice north-south and east-west orientation of the walls, though possible naturally, is a strong indicator that humans were involved. Finally there is that ancient local legend about a "castle" that vanished beneath the sea…
But here, as everywhere else, more research – much more research – is needed to settle the matter.