(December 2009) Forthcoming in Skeptical Adversaria or Skeptical Intelligencer
Reproduced with permission
In his books 1421 and 1434, Gavin Menzies has promoted the idea that Chinese navigators explored much of the globe in the 15th Century, leaving ample evidence – including linguistic evidence – which has been ignored by mainstream historians. His views have achieved considerable popular success (partly because of his adroit self-promotion) but have not found favour in the scholarly world. Menzies regards his academic critics as ill-informed, biased and even mentally ill, and refuses to engage in exchanges with them (he rejected my own attempts at dialogue).
Menzies’ linguistic material in 1421 is mainly located in his ‘Appendix 1’ (pp 494-595), from which all but a few of the examples cited below are drawn. (All page references are to the 2003 Bantam paperback edition, IBSN 0 553 81522 9.) For the most part, the treatment of each claim given in the appendix is very brief indeed. Many of the sources are badly dated and reflect ideas which have long been superseded. In some cases, eg the list of items on p 518, no source is provided.
The problems with Menzies’ linguistic material include the following:
1) Menzies’ philological methods are covert, but his equations of Chinese and non-Chinese words suggest strongly that they are typical of the amateur fringe (and similar to those which were used by scholars too before the subject had been properly investigated). Such methods involve the unjustifiable assumption that unsystematically and/or loosely similar forms with similar (or allegedly connected) meanings, in languages not necessarily known to be related ‘genetically’ or to have been in contact at relevant dates, are probably connected – either because the languages are in fact ‘genetically’ related, or (in the vast majority of the cases adduced by Menzies) because of influential contact between their speakers (linguistic transfer or ‘borrowing’) – in this instance, the influence of Chinese upon other languages. It is imagined that such pairs/sets of similar forms can thus be used to strengthen an argument to the effect that the cultures in question were connected in either of these two ways.
However: there are millions of words and word-parts in thousands of languages, and there are only so many common sounds and sound combinations. Superficial similarity between words and/or word-parts taken from different languages, especially short ones, is in itself no evidence of a genuine connection, even if the meanings are similar. If the meanings are not especially similar, or are merely alleged to be related as part of some theory, the case is even weaker. The upshot of this is that the onus has to lie upon those who present novel etymologies or claims about cognatehood (shared origin) to show that there is at least a good case. (Indeed, for recent times where hard evidence is plentiful, there are very many superficially similar forms, often with similar meanings and/or in related languages, which are nevertheless demonstrably unconnected and only accidentally similar.)
In the course of 200 years of intensive study, linguists have learned much about the strength of evidence that is required to demonstrate (probable) connectedness between such forms. The main issue is that of systematicity; language change is very largely systematic. It is almost universally agreed that sets of unsystematically similar forms with similar meanings are not at all likely to be cognates, and there is certainly no reason to regard them as demonstrably cognate. This constraint applies most strongly in cases of alleged ‘genetic’ relationship, but it still applies to cases of alleged contact as well.
Anyone who wishes to overturn these points needs to develop arguments to the effect that the scholarly tradition of historical linguistics is mistaken in these respects. Menzies makes no attempt to do this, or to justify his linguistic equations within this constraint established by the tradition.
Another issue here involves known or very well-grounded established etymologies for words, involving the histories of the relevant language families. Proposers of alternative etymologies need to argue that these are more plausible than the established ones.
Any major loosening of the standards of evidence for cognatehood, which sets of claims such as Menzies’ require, would have the consequence that very many alternative proposals (involving eg a whole range of different languages of origin for the same words) would be roughly equally plausible. But these proposals all contradict each other; only one of them, if any, could be correct. (In that event, the reasonable conclusion would probably be that we could not say much at all about philology or older etymologies with any confidence. Orthodox linguists would regard this conclusion as a last resort and as not warranted by the actual evidence.)
Like most non-linguists who advance such theories, Menzies concerns himself only with vocabulary, saying nothing about phonological or grammatical similarities/differences between Chinese and the non-Chinese languages in question. Since his claims involve contact rather than ‘genetic’ relatedness, this omission is less serious than it would be if the latter were at stake; but his total failure to deal with these aspects of the languages remains potentially problematic.
Examples of Menzies’ philological conclusions follow. All of them involve unsystematic and/or phonetically loose similarities, and in some cases (eg the list of items on p 518) much of the information needed for assessment is in fact missing. In a few cases, no actual forms at all are cited.
95 allegedly Chinese words and 130 allegedly Chinese placenames (not intelligible as Aymara, Quechua etc) in northern Peru (pp 517-518); 15 of these words are listed on p 518 (as noted, without any reference), but the actual forms of the allegedly connected Chinese words are not themselves provided for comparison with the Peruvian words, only their meanings are given – and the alleged relevance of these meanings is not made apparent, since no descriptions of the locations are offered.
Country names Peru and Chile interpreted as Chinese (pp 517, 565); the meanings of the allegedly connected Chinese words are only very loosely relevant.
Fifteen further words/phrases involving allegedly significant similarities between Chinese and non-Asian languages (or in some cases between Asian languages other than Chinese and non-Asian languages, with the relevance to Chinese being left unexplicated), with an unexpanded reference to the existence of at least 36 more (and an abrupt reference to the genuinely interesting case of Polynesian kumara/umara (‘sweet potato’) and phonetically similar words in Peruvian languages referring to this vegetable, which clearly did spread across the Pacific in pre-modern times in some manner, though there seems no reason to posit Chinese involvement) (pp 518-519); the source for one of the 15 items is the fringe writer Henrietta Mertz, and some items are again unreferenced.
Reference to Nancy Yaw Davis’ near-fringe methodologically unsound claims regarding significant similarities between Zuni and Japanese (p 518; see also 2) below); again, the relevance to Chinese is left unexplicated, and in this case specific words are not mentioned.
Reference to Columbus’ reported meeting with people calling themselves Yin , identified as the Chinese word for ‘India’ (p 553); similarity between such short word-forms is of no significance, since, as noted, it could very readily arise by chance.
Claim (presumably associated with the foregoing) that the word Inuit derives from Yin Uit (‘Chinese’ for ‘people from India’), attributed (like some other such points) to Martin Tai (p 558).
Unreferenced claim that the Inca ruler-name Atahuallpa is a Chinese word for ‘chicken’ (p 563).
Claim (again attributed to Martin Tai) that the name Inca itself involves the Chinese word yin (p 565).
Menzies also makes some extremely vague references to ‘American Indian names that are Chinese’ (p 519), to linguistic similarities between Aboriginal languages and Japanese (p 525; no reference; relevance, again?), to ‘language’ in the context of alleged Chinese cultural exports to California (p 546), to linguistic similarities between Peru and China (p 564), and to ‘Chinese linguistics’ (a very odd term in context!) on Niue (p 574).
In none of these cases is there any reference in Menzies’ text to current mainstream linguistic scholarship. The languages in question have been researched adequately or better, and if any such cases had any empirical justification they would be so sensational as to require discussion.
2) Menzies adduces some cases involving the allegedly surprising use of written Chinese or other (semi-)relevant languages in areas remote from Chinese influence as understood by scholars, or of other Asian languages found written in areas remote from East Asia. Some of these cases (eg that of the ‘Tamil Bell’ found in New Zealand; see pp 210-212, 517) are genuinely interesting but are not evidence of Chinese activity as posited by Menzies. Others, eg the alleged Mongolian inscription on Dighton Rock in Massachusetts (p 555) or the alleged Chinese and Mongolian inscriptions on the wall of a temple in Mexico (p 562), would clearly be relevant (and very dramatic) if verified but are not in fact accepted as accurate by any qualified scholar. Still others relate to events so remote in time and involving such perishable materials that they very probably could not now be tested (eg the claim on p 563 that the Spanish conquistadores found Mongolian script used on paper in Mexico, or the claims on p 564 that Chinese script was found on pots and mummy-cases in early colonial Peru). Yet others are too vague to be of use to Menzies. There is no reason to suppose that an unintelligible inscription allegedly found (when?) in the Azores by the Portuguese (p 561) was in Chinese (or in any other relevant language).
3) Menzies adduces various cases allegedly demonstrating the surprising use of spoken Chinese in areas normally regarded as remote from Chinese influence, or the surprising mutual intelligibility of Chinese and a range of non-Asian languages. Examples:
Two Chinese-speaking/understanding villages in Peru (pp 514, 517, 564; allegedly, the inhabitants of these two villages do not understand each others’ local ‘dialects’!) Chinese-speaking location in California (p 517) 20th Century Navajo-speakers understanding Chinese (pp 559-560) Zuni-speakers understanding Japanese (p 559); Davis again and associates (relevance?)
In none of these cases is there any reference to current mainstream linguistic scholarship. Again, the languages in question have been researched adequately or better, and if any such cases had any empirical justification they would be so sensational as to require discussion. Many of these cases were reported so long ago that they could not now be disconfirmed; negative findings today could be met with the rejoinder that matters are now different.
In respect of mutual intelligibility: some ‘epigraphist’ writers such as Cyclone Covey (a supporter of the late Barry Fell) have recently endorsed cases of this kind in support of other extreme diffusionist linguistic claims relating to alleged pre-Columbian transatlantic or transpacific voyages. Covey’s leading case involves one of the languages mentioned in this context by Menzies: early-mid 20th Century Navajo. This time, though, the other language involved is not Chinese but Uighur (Turkestan). Covey believes (with Ethel Stewart and others) that some (non-Inuit) Amerindian groups such as the Navajo actually left Central Asia only in the last 1000-3000 years. The Navajo migrated in medieval times, fleeing the Mongols. Their language is therefore still close to the Uighur spoken by those who remained in Asia. Obviously, either Menzies (and his source John Ting) or Covey must be wrong, as Uighur and Chinese are not similar; very probably, both are wrong, as Navajo is very unlike both Uighur and Chinese.
Fringe works down the years (especially older works) have reported many other such incidents – Irish Gaelic understood by Mexican Amerindians, Latvian by Tatars, Welsh by speakers of Mandan in the American Mid-West, etc, etc – but (not surprisingly) the accounts are anecdotal only and actual evidence is never forthcoming. Most such reports are clearly apocryphal. (Even Breton and Welsh, closely related P-Celtic languages for which such claims are made, are not now mutually intelligible apart from the odd phrase.)
Covey himself thinks, bizarrely, that it is up to the linguistics establishment to test such claims, otherwise, they should apparently be regarded as reliable. (In any case, negative findings on the Navajo-Uighur case, obtained today, would be countered by Covey’s claim – made in correspondence with me – that matters are now different.)
4) In his linguistic equations, Menzies typically does not identify the type of spoken Chinese in question. Because Mandarin and the various fangyan (‘dialects’) such as Cantonese, Hokkien, Teochew etc spoken around China have rather divergent phonologies and partly separate lexical phonologies, there are usually major phonetic and phonological differences between the various spoken forms of any given Chinese word written throughout China with the same character and regarded by the Chinese as the same word with the same meaning. Phonological change in Chinese (of any type) over the centuries (even recent centuries) is also potentially a factor. Menzies ignores all this complexity, citing forms simply as ‘Chinese’. This potentially allows him considerable freedom in looking for a ‘Chinese’ word displaying (often loose) phonetic similarity with a given non-Chinese word.
5) Some of the ‘Chinese’ words cited have unlikely phonological forms. Most notably, none of the major fangyan , nor indeed Mandarin, possess words terminating in /-l/; but Menzies cites the form tsil (‘wet’) (pp 518, 557) as a source for a word spelled identically in roman script and attributed to the Squamish of British Columbia. He is under a clear obligation to identify the varieties of Chinese in which he claims such forms occur.
6) Menzies’ other linguistic statements are sometimes inaccurate. One involves another case of loose research methodology; it relates to Menzies’ means of ‘identifying’ the script on stones found in the Cape Verde Islands and in the Congo (pp 134-136). Menzies thought that it ‘looked Indian’, but rather than take the obvious route of consulting a university linguistics or Indology department he asked the Bank of India. Their employee advised him that it ‘look[ed] like Malayalam’, and on further enquiry told him that this was (my emphasis) the language of Kerala. That is correct, but Malayalam is far from dead, as was might imply; it still has many millions of speakers in India and in a substantial diaspora. And, whether the information about Malayalam was wholly accurate or not, Menzies could readily have found experts in the language; as things actually are, he could readily have found literate native speakers as well. This is not as bad as Erich von Daniken’s disgracefully lazy decision to quote Exodus from memory in print rather than looking up the relevant passage – but it is not much better! (Even if Malayalam is involved here, of course, that is not itself evidence of Chinese activity as posited by Menzies.)
Menzies’ further linguistic errors include the claim that Haida and Aleut are the same language (p 558).