Interlinguistics / Interlingüística / Interlinguistik / Interlingvistiko
How European is Esperanto? A typological study* Mikael Parkvall
Stockholm University
The typological similarities between Esperanto and other languages have long been a matter of debate. Assuming that foreign-language structures are more easily acquired when they resemble those of the learner’s native tongue, any candidate for a global lingua franca obviously ought to be as typologically neutral as possible. One common criticism of Esperanto is that it is ‘too European,’ and thus less accessible to speakers of non-European languages. In order to provide a more solid base for such discussions, this paper makes an attempt to quantify the Eurocentricity of Esperanto, employing the features catalogued in the World Atlas of Language Structures. It is concluded that Esperanto is indeed somewhat European in character, but considerably less so than the European languages themselves. Keywords: Esperanto, interlinguistics, typology, language structure, syntax, phonology, comparative linguistics, Eurocentrism
Since the birth of Esperanto and the Esperanto movement, there has been a continuous debate on the advantages and disadvantages of promoting Esperanto as the main international language. One of the focal points of this discussion has been the language’s typological neutrality, or lack thereof. Sceptics have emphasised that Esperanto is essentially European in nature, and therefore presumably less easily learnt by non-Europeans. Proponents, not unexpectedly, have tended to downplay the similarities between Esperanto and European languages, and instead preferred to emphasize the logical and exceptionless characteristics of the grammar. The arguments from both sides have essentially rested on gut feeling and the comparison of bits and pieces of grammar, but to the best of my knowledge, no truly systematic comparison has ever been made. The publication of the World Atlas of Language Structures (Haspelmath 2005, henceforth WALS) allows us to perform precisely such a comparison for the first time, and the aim of this paper is to quantify the degree to which (if at all) Esperanto is a Euro-centric language.
S F O PRO
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B
Language Problems & Language Planning 34:1 (2010), 63–79. doi 10.1075/lplp.34.1.04par issn 0272–2690 / e-issn 1569–9889 © John Benjamins Publishing Company
©J
1st proofs
64 Mikael Parkvall
The data WALS is the largest typological database of human languages ever compiled. It contains 142 chapters, each dealing with a specific linguistic feature. Between 113 and 1,371 languages are classified with regard to these features. Some chapters use a binary distinction, while others have divided the world’s languages into three to nine categories. Combining the 142 linguistic features, WALS includes more than 2,500 languages, but not all of these are equally well represented in the atlas. The values for languages such as English and French are included for virtually all features, but many less well-documented non-European tongues are only sporadically featured. The total number of data points, therefore, is not 2,500•142 ≈ 355,000, but rather about 60,000. Now, by entering the corresponding values for Esperanto (which is entirely absent from the WALS database), we can finally make a comparison between Esperanto and a large (and representative) number of other languages that spans all areas of linguistic structure. Five WALS chapters are excluded from consideration here: 139–140 (because they deal with signed languages, which Esperanto is not), 141 (because it treats writing systems on a country rather than a language basis), and 77–78 (because I simply failed to completely understand the classification used). All in all, then, the material used here consists of 137 linguistic features, each with two to nine possible feature values. A final note on the data used: Not only is WALS the most inclusive typological resource ever, but it also has the additional advantage of being compiled by the world’s most renowned typologists without any specific reference to Esperanto. While my interpretation of the results could be criticised in various ways, it is nonetheless the case that the comparandum relies on judgements made by linguists with no particular interest in the issue discussed here (after all, none of them included Esperanto in their sample). While this absence guarantees the neutrality of the data, it does not automatically imply that we are dealing with the perfect data set: semantics, for instance, is under-represented in WALS, and there is a possibility that this skews the overall results.
Method Since Esperanto is not featured in WALS at all, the first task was to enter the relevant values for this language into the database. As my own active competence is somewhat limited, I elicited example sentences from two highly competent speakers of Esperanto,1 who in general provided highly convergent answers. The respondents were explicitly instructed not just to provide possible translations, but
S F O PRO
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
How European is Esperanto?
65
rather the translations they felt would be the most natural counterparts of the example sentences. This is of some importance since Esperanto (like presumably most other languages) permits certain alternatives that are only rarely exploited by speakers. For instance, while it allows for free word order, SVO is by far the most frequently occurring. The same thing applies to, for example, Russian, which in WALS is classified as an SVO language, and allowing only the unmarked option in this study is thus essential in order to ensure comparability. Despite the options present, Esperanto is therefore treated as SVO in my comparison. The main problem, however, was not to figure out how Esperanto does this or that, but rather to understand the exact details behind the existing classification. Each WALS chapter contains a few pages of text where each author outlines their guiding principles, but it is of course difficult to cover all potential problems. While some features were rather straightforward, others required a considerable effort to “get into the head” of the author.2 Some errors no doubt remain, but (with the exception of Chapters 77–78, which were excluded for precisely the reason that I could not grasp the basis of the classification well enough to apply it with any confidence), I hope and believe that Esperanto is now relatively correctly represented.
Comparison We can now compare the degree of similarity between Esperanto and any other language or group of languages, provided they are featured in WALS. I first made a comparison based on individual languages. As already mentioned, WALS contains data on about 2,500 languages, but most of these are only very sporadically featured, and can therefore not be used for comparisons beyond one feature or a handful of features. I chose to consider only languages with data for 76 or more of the 137 features.3 This gives us a total of 180 languages with which to compare Esperanto. The average language makes the same choices as Esperanto in 44.3% of all cases, i. e. it performs identically to Esperanto on 61 of the 137 features. The range is between 32.6% and 69.9%, so, in other words, any given language has between one third and two thirds of its structure in common with Esperanto. As can be seen in Table 1, the nine languages the most similar to Esperanto are all Indo-European languages of Europe, and with the exception of Irish (rank 29), these are in fact the only such languages of the 180 considered here. This does indeed convey a picture of Esperanto as a rather Eurocentric language, although it is interesting to note that ranks 11–20 include languages which are neither Indo-European nor European, such as Brahui, Khasi, Quechua,
S F O PRO
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
66 Mikael Parkvall
Table 1. The ten languages most similar to Esperanto Language
No. of features considered
% similar to Esperanto
1.
Russian
133
69.9%
2.
English
137
67.2%
French
134
67.2%
Albanian
76
67.1%
4.
Polish
76
67.1%
6.
German
127
64.6%
7.
Greek (modern)
127
63.8%
8.
Latvian
110
63.6%
9.
Spanish
133
60.9%
10.
Finnish
133
58.6%
Kannada, Drehu, Malagasy and Korean. All of these are more similar to Esperanto than the average language (44.3%). The ten languages which are the least similar to Esperanto (32.6%–34.7%) are all spoken in Australia or the Americas. With regard to groups of languages, a few options stand out as particularly worthy of closer examination. First, given the claim that Esperanto is heavily European in its structure, we would like to compare it to European languages in general. However, some European languages are normally considered “more European” than others, and the label Standard Average European (SAE) was coined in 1941 by Benjamin Whorf (1956: 25) for the core of languages sharing certain key features. In the following, SAE is taken to consist of all Germanic and Romance languages, as well as Western and Southern Slavic and Balkan languages (with Dutch, German, French and Italian assigned twice their weight; cf. van der Auwera [1998] and Haspelmath [1998]). Third, Romance languages deserve special attention for the reason that the majority of the Esperanto lexicon is derived from them (Janton 1993: 51), leading to the common layman’s view that Esperanto is “some kind of Spanish” or “very much like Italian.” Fourth, we might expect Esperanto to display strong similarities to languages known by its creator, L.L. Zamenhof. Details vary, but according to standard biographies, these would seem to have included Polish, Russian, Yiddish, German, French, Latin, classical Greek, ancient Hebrew and English.4 Finally, it is of obvious interest to compare Esperanto with the languages of the world as a whole. Applying the same technique5 to groups of languages as was just done with individual varieties, we get the following degrees of similarity:
S F O PRO
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
How European is Esperanto?
67
Table 2. Degree of similarity between Esperanto and various groups of languages7 % agreement with Esperanto Known by Zamenhof I
78.1%
Known by Zamenhof II6
76.6%
European languages
75.2%
SAE
74.5%
Romance languages
70.8%
Languages of Asia
56.8%
Languages of the world as a whole
54.0%
Languages of North America
53.0%
Languages of Africa
51.9%
Languages of South America
48.9%
Languages of Oceania
46.3%
These figures again suggest that Esperanto is indeed relatively close to those languages that it could be suspected to resemble, namely European languages in general, and in particular those known to its creator. Some other groups of languages have also been suggested to be remarkably close to Esperanto. Piron (1981), for instance, argues for a special relationship between Esperanto and Asian languages. However, my scepticism of his claim is supported by the fact that the language he especially favours, Mandarin, is in fact no more similar to Esperanto (44.5%) with regard to the traits examined here than the average language of the world is. The same author also claims that “at the middle plane [=syntax] Esperanto is indubitably Slavic”. While the overall similarity between Esperanto and Slavic is 73.3%, it is 81.5% in the realm of syntax,8 so Piron’s second statement seems indeed to have some merit.
But then again… So, according to the method pursued here, Esperanto is indeed a European language. But that is not necessarily the end of the story. We could also say that despite a general European-ness, in more than half of all WALS features (74 of 137), Esperanto has chosen whatever strategy happens to be the most common worldwide. That is, while the average individual language has a mere 44.3% similarity with Esperanto, the parameter setting displayed by the artificial language is the most common one world-wide in 54.0% of the cases.
S F O PRO
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
68 Mikael Parkvall
In this context, one may wonder whether the 54% similarity between Esperanto and the preferred world-wide option is high or low. It turns out that in this metric, Daga (a language of New Guinea) is “the world’s most representative language,” in that it picks the cross-linguistically most common option in 70.8% of the features considered.9 At the other end of the scale is the Amazonian language Wari’, which adheres to the standard in only 43.4% of all WALS features. The average language (both in terms of mean and median) scores 55.4% here, while Esperanto, as just mentioned, ends up with 54.0% similarities. In other words, Esperanto is slightly less like the average language, than, as it were, the average language is. Among the 181 languages examined here, Esperanto ranks 110 in terms of agreement with the world’s most common parameter settings. Now, with Esperanto ranked 110 out of 181 on the “normality” list, the (other) European languages fare far less well. Of the nine Indo-European languages in Table 1, for instance, all rank lower, in many cases by quite some margin. German, for instance, is the 178th most “normal” language (out of 181!), that is, one of the most “exotic” there are. We could interpret this as meaning that Esperanto is a slightly European language, but considerably less so than the European languages themselves. So, it is in a minority of cases that Esperanto has opted for another strategy than the most common one world-wide. This is usually — but by no means always — one that predominates in Europe. Table 3 sets out the wals features according to whether the value for Esperanto is the preferred one in the world as a whole, or only in Europe, or neither. Table 3. The feature values in Esperanto according to whether these match the preferred value in the world as a whole, in Europe (but not in the world as a whole), or neither no. of features
%
wals maps
World-wide features
74
54%
1, 2, 5, 6, 7, 8, 9, 10, 11, 13, 15, 16, 18, 19, 20, 24, 26, 28, 33, 34, 35, 37, 39, 40, 42, 43, 45, 48, 51, 52, 55, 58, 59, 60, 61, 63, 64, 65, 66, 68, 69, 73, 74, 79, 80, 82, 83, 84, 88, 90, 91, 93, 94, 96, 100, 104, 105, 106, 108, 109, 111, 112, 114, 116, 117, 120, 124, 125, 128, 129, 130, 131, 135, 137
European features
44
32%
4, 12, 17, 23, 25, 27, 30, 31, 36, 41, 44, 47, 54, 56, 57, 62, 71, 75, 81, 85, 86, 87, 89, 92, 95, 97, 98, 99, 101, 107, 110, 113, 115, 118, 119, 121, 122, 123, 126, 127, 132, 133, 134, 138
Neither European nor 18 world-wide
13%
3, 14, 21, 22, 29, 32, 38, 46, 49, 50, 53, 67, 70, 72, 76, 102, 103, 142
S F O PRO
D E ANY T P M C O O R R E LISHING C
TOTAL
136
100%
U ONH NCB E N J A M I N S P U B ©J
1st proofs
How European is Esperanto? 69
Manifestations of Eurocentricity European-ness is of course a relative concept. In some of the cases classified as “European” above, the value for Esperanto matches the second most common one in the world, and this is sometimes second only by a narrow margin. Such instances cannot count as very strong evidence for the European character of Esperanto. Others, however, must be considered very European. In the case of seven features (28, 54, 62, 79, 106, 123, 133) where Esperanto has settled on a “European” value, there is a negative correlation between the value’s frequency in Europe and in the world as a whole. We could measure this Eurocentrism in a variety of ways, but I shall refrain from presenting a ranking and simply point out some features that stand out as conspicuously European (here meaning “Indo-European of Europe”) in character. – Esperanto has, like European languages, a large number of colour terms, and in particular, it makes a distinction between ‘green’ and ‘blue’. – Esperanto distinguishes between ‘he’ (li) and ‘she’ (ŝi), unlike most languages outside Europe. – Most non-European languages make use of reduplication as a morphological device (corresponding to non-existing English constructions such as *run-run ‘to run fast’ or *big-big ‘huge’).10 Esperanto aligns with European languages in not exploiting this possibility. In particular, for distributives non-European languages have a preference for constructions of the type *four-four ‘four each’, where Esperanto uses a more (eastern) European po kvar (with po being a preposition-like element)11 for the same meaning. – It is common for non-European languages to have an associative plural, often of the type *the Pauls ‘Paul and his friends/colleagues/relatives.’ Esperanto, like many European languages, lacks this feature. – Esperanto comparatives function like European ones, i. e. with a comparative particle: Mi estas pli juna ol vi ‘I am younger than you.’ In the rest of the world, the most common strategy is the locational one, which would translate as *I am younger from/on/to you. – The use of a relative pronoun in a sentence such as La viro kiu salutis min estis germano ‘The man who greeted me was a German’ is highly European. Most languages would simply leave out the morpheme corresponding to the who here.
S F O PRO
One area which is barely featured in WALS, but which would seem to contain more Europeanisms, is that of semantics. Comrie (1996: 40) points out, for instance, that Esperanto merges ‘carry’ and ‘wear’ (porti) in the way many European languages do (although many speakers would prefer surhavi in the sense of ‘wear’). There are
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
70 Mikael Parkvall
also words such as stelklara ‘starry’, vidpunkto ‘viewpoint’, mejloŝtono ‘milestone’, terpomo ‘potato’, lav-urso ‘racoon’, eldoni ‘to publish’ and plenkreska ‘fully grown’ which appear to be calqued directly on European patterns.
Non-European and typologically unusual structures In some cases, Esperanto is patently un-European, and at times it parallels the extra-European preference. A good example of this is feature 52, which demonstrates that the merger of comitatives and instrumentals is something that primarily European languages engage in (cf. I went to the cinema with John ~ I cut the bread with a knife). Esperanto, however, sides with the majority option in the rest of the world in keeping the two apart (instrumental preposition per, comitative preposition kun). It is possible that Zamenhof was inspired by Polish or Russian in this case, since those languages are equally un-European despite their geographical location. Regardless of the historical explanation, the fact remains that Esperanto in this case has made a “typologically sound” choice. Perhaps the most interesting feature in the entire sample concerns feature 46, where Esperanto has a value not attested in any natural language. While most of the world’s languages have indefinite pronouns (‘somewhere’, ‘something’, ‘somebody’, etc.) based on interrogatives (‘where’, ‘what’, ‘who’, etc.), Esperanto presents exactly the opposite picture: the interrogatives kie, kio and kiu are based on the indefinites ie, io, iu. This is interesting especially in the light of first-language acquisition studies. While descriptions of natively spoken Esperanto (Bergen 2001; Corsetti, Pinto & Tolomeo 2004; Lindstedt 2006; Versteegh 1984; Versteegh 1993) have pointed out some discrepancies between this and the prescriptive norm, this particular feature has never been reported to cause any troubles in first language acquisition. In other words, the Esperanto strategy is compatible with the human language faculty, even though it has not been documented in any non-artificial language.12
Does “normal” equal good? The tacit assumption thus far, both in this paper and in other writings, has been that “normal is good.” The rationale behind this assumption is that a linguistic feature is taken to be easier to acquire if it matches one’s own mother tongue. The “neither European nor world-wide” row in Table 3 contains features which are relatively unusual, but this does not necessarily imply that they are dysfunctional — in particular in view of Esperanto’s intended role as an auxiliary second language.
S F O PRO
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
How European is Esperanto?
In order to be a perfectly representative language, Esperanto would have suppletive morphology and portmanteau morphs, and yet, surely no one would consider the addition of these features an improvement, since it would increase the burden both of learning and of processing. Similarly, the most common argument marking is one where both the agent and the patient are morphologically marked on the verb, but despite all the reform proposals, few (if any) have advocated that Esperanto would benefit from such an arrangement. Most languages also have possessive affixes, so that the possessor is affixed to the possessum (*my-arm or *arm-my ‘my arm’), but again this would not necessarily increase the acquisitional ease, as analytical structures are often considered (correctly or not) more easily learnt than synthetical ones.13 In order to be less European, Esperanto could be equipped with phonemic tone, as in most African and East Asian languages. No doubt, however, this would attract a good deal of criticism. In other words, it is not particularly difficult to come up with features which Esperanto does display (or lack), but which can be considered “European” or typologically unrepresentative, but where the alternative option is likely to stir up more criticism than the present one ever has. In short, there are plenty of cases where alleged simplicity or learnability clashes with typological frequency. In a sense, this is hardly surprising, since most languages have in all likelihood had a relatively small proportion of second-language learners among their speakers for most of their history, and they have thus experienced only a moderate evolutionary pressure to become optimised for second language acquisition. Other features are less obviously well chosen. To agree more with the makeup of the average language, Esperanto would, for instance, lack case-marking and passivisation, and favour SOV word order and postpositions. If we were to construct a language from scratch, these could be considered better choices than their counterparts in Esperanto, on the basis of their higher frequency in attested human languages. It could be argued that passivisation, for example, is merely an option that the speakers can make use of, but that they need not employ (though the listener would of course have to be able to decode it). The phonology, however, cannot be avoided, and both from a segmental and a phonotactic point of view, I see its phonology as one of the least fortunate parts of Esperanto, with its profusion of consonants and the occurrence of complex clusters (as in, for example, ŝraŭbo [ʃrawbo] ‘screw’, vrako ‘wreck’, skvamo ‘scale of fish’, and, not least, scienco [stsientso] ‘science’) that lack counterparts in many natural languages.
S F O PRO
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
71
72
Mikael Parkvall
Taking speaker populations into account One person who read an earlier draft of this paper suggested that numbers of speakers be taken into account. Instead of counting the number of languages preferring one or another strategy, one could count the number of speakers accustomed (or otherwise) to that strategy. I have chosen not to include such a mathematical exercise here, but the result is nevertheless clear, in that it would make Esperanto look considerably more in line with the global state of affairs. That result, in turn, could be amplified by counting not only mother-tongue speakers, but also including second language users. The simple reason for this outcome is that the European languages are typically spoken by vastly greater communities than are the native languages of Australia or the Americas.14 This alternative perspective, naturally, would not prove that Esperanto is less (or more) European than I have portrayed it here, but merely that the world’s population is linguistically more European than the above analysis tends to imply. Also, since English — for the time being the only realistic alternative to Esperanto as a global lingua franca — is the most widely spoken language if L2 speakers are included, the suggested modus operandi could plausibly lead to English emerging as a language at least as neutral as Esperanto, in terms of its structural relationship to languages already known and used by the world’s population. Interesting as such an comparison might be, however, it falls beyond the scope of this paper.
Concluding remarks Though I think it is always useful and desirable to quantify existing gut feelings, the main conclusion from the above is probably dependent on one’s own ideological position. Esperanto is indeed more European in character than many of its advocates would have it, but probably less so than many of its opponents would have predicted (in particular those with only a fleeting acquaintance with its lexicon). In other words, in terms of Esperanto’s structural affinities with the known range of human linguistic potential, the glass could be seen as either half full or half empty. It may be worth bearing in mind that most Esperanto texts in history have probably been produced by people with a European mother tongue, and, in addition, this also applies to my two informants. In this context, Lindstedt (2006: 48) points out that Esperanto norms are far more dependent on speakers (as opposed to active language planning) than most people think. It might therefore be interesting to study the differences between the Esperanto portrayed here and that used by people without knowledge of a European language. There is an obvious
S F O PRO
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
How European is Esperanto?
73
possibility that such versions would emerge as less Eurocentric. As the language is created by its speakers, a possible shift in speaker demographics could perhaps lead to a drift away from European-influenced norms.
Notes * My thanks go to Päivi Juvonen, Sonja Petrović Lundberg, Hartmut Traunmüller, Joakim Enwall, Bertil Wennergren and two anonymous reviewers for their constructive input. I also owe special credit to Håkan Lundberg, whose questions provided the initial impetus for this article. 1. Native speakers of German and Swedish respectively; both use Esperanto as their home language with their spouses who are of different linguistic backgrounds. 2. Contacts with some of the WALS contributors provided some help here. 3. Any such cut-off point is bound to be arbitrary, of course. In this case, I picked 76 simply in order to include Polish, a language of special interest here because it was spoken by Esperanto’s creator. 4. WALS does not include extinct languages, and therefore excludes Latin, and only features the modern versions of Greek and Hebrew. It also has extremely few mentions of Yiddish, which was one of the languages Zamenhof grew up speaking. 5. I am here using the majority option within each group as the group value. 6. This figure includes the modern versions of Greek and Hebrew as stand-ins for the classical varieties. 7. It should be borne in mind here that labels such as “American languages” and “Oceanic languages” (as is common practice in linguistic typology) refer to speech varieties indigenous to the respective continents. While the Americas and Oceania are nowadays completely dominated by (originally) European languages, these are still counted as European, while only the pre-1492 languages are treated as ”American.” 8. Here defined as WALS features 58–64 and 81–128. 9. Note that this refers only to the features for which there are any data in WALS. Also, as already mentioned, I included languages with 76 or more attestations in the 137-feature database. For Daga, for instance, 89 values are included. It is of course possible that Daga behaves excentrically with regard to the 48 features about which WALS is silent. 10. Here and in the following, English morphemes are used only to illustrate the types of constructions concerned. For readers with a primarily passive command of English, their ungrammatical nature is marked by means of an asterisk.
S F O PRO
11. One of the anonymous reviewers points out that there is some variability among speakers of Esperanto here, and at least for many users, po fails to display the characteristuics of a preposition proper.
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
74
Mikael Parkvall 12. Note, however, that one of the anonymous peer reviewers has reservations about my account here. S/he writes (originally in Esperanto) that “Instead of k-iu one could also propose ki-u (compare the coinages alies, aliam, and not e.g. ali-ies: such forms must contain an -i-, but it is felt to belong to the first, not the second part of the word.” Moreover, the reviewer reports on two native speakers that they “misspell ne niu as two words, but never nen iu”. 13. According to a colleague who is more familiar with the second language acquisition literature than I, this claim appears not to have been empirically tested. 14. Even Esperanto itself must be considered a larger-than-average language when L2 speakers are included.
References Bergen, Benjamin. 2001. Nativization processes in L1 Esperanto. Journal of Child Language 28/3:575–595. Comrie, Bernard. 1996. Natural and artificial international languages: A typologist’s assessment. Journal of Universal Language 3:35–55. Corsetti, Renato, Maria Antonietta Pinto & Maria Tolomeo. 2004. Regularizing the regular: The phenomenon of overregularization in Esperanto-speaking children. Language Problems and Language Planning 28/3:261–282. Haspelmath, Martin. 1998. How young is Standard Average European? Language Sciences 20/3:271–287. Haspelmath, Martin, Matthew Dryer, David Gil & Bernard Comrie, ed. 2005. The World Atlas of Language Structures. Oxford: Oxford U. P. Janton, Pierre. 1993. Esperanto: Language, Literature and Community. Albany: State U. of New York Press. Lindstedt, Jouko. 2006. Native Esperanto as a test case for natural language. SKY 19:47–55. Piron, Claude. 1981. Esperanto: European or Asiatic Language? Esperanto Documents 22A. Rotterdam: Universal Esperanto Association. Van der Auwera, Johan. 1998. Revisiting the Balkan and Meso-American linguistic areas. Language Sciences 20/3:259–270. Versteegh, Kees. 1984. Piĝinigado, kreoligado kaj Esperanto. Hungara Vivo 24/4:127–129. Versteegh, Kees. 1993. Esperanto as a first language: Language acquisition with a restricted input. Linguistics 31/3:539–555. Whorf, Benjamin. 1956. Language, Thought, and Reality. Selected Writings of Benjamin Lee Whorf. Cambridge: MIT Press.
Sammanfattning Hur europeiskt är esperanto? En typologisk studie
S F O PRO
De typologiska likheterna mellan esperanto och andra språk har länge debatterats. Antagandes att drag från ett främmande språk lärs in lättare om de uppvisar
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
How European is Esperanto?
75
likheter med det egna modermålet, borde givetvis ett internationellt hjälpspråk vara så typologiskt neutralt som möjligt. Ett vanligt stycke kritik mot esperanto är att det är ”alltför europeiskt”, och därmed mindre tillgängligt för talare av utomeuropeiska språk. I syfte att erbjuda en bättre grund för sådana diskussioner försöker denna artikel — med hjälp av de språkdrag som listas i World Atlas of Language Structures — kvantifiera den eventuella eurocentriskheten hos esperanto. Slutsatsen är att esperanto mycket riktigt är en smula europeiskt till sin karaktär, dock i betydligt mindre utsträckning än de europeiska språken själva.
Resumo Kiel eŭropa estas Esperanto? Tipologia studo La tipologiaj similecoj inter Esperanto kaj aliaj lingvoj estas delonge temo de debatoj. Se oni supozas, ke la strukturoj de fremda lingvo estas des pli facile akireblaj, ju pli ili similas al tiuj de la denaska lingvo de la lernanto, tiam evidente ĉiu kandidato por la rolo de monda interlingvo devus esti tipologie kiel eble plej neŭtrala. Ofte oni kritikas Esperanton pro tio, ke ĝi estas “tro eŭropa”, kaj tial malpli facile alproprigebla por neeŭroplingvanoj. Cele al kreo de pli firma bazo por tiaj diskutoj, ĉi tiu artikolo provas mezuri la eŭropecon de Esperanto, utiligante la trajtojn katalogitajn en World Atlas of Language Structures. Oni venas al konkludo, ke Esperanto ja estas iom eŭropeca, sed atentinde malpli tia ol la eŭropaj lingvoj mem.
Appendix: The Wals features As explained earlier, five chapters were excluded from consideration here. These are bracketed in the following table. The labels are not necessarily self-explanatory, but it would take up disproportionate amounts of space to define them here. The interested reader is therefore referred to Wals itself for precise definitions. 1 Consonant Inventories 2 Vowel Quality Inventories 3 Consonant-Vowel Ratio 4 Voicing in Plosives and Fricatives 5 Voicing and Gaps in Plosive Systems 6 Uvular Consonants 7 Glottalized Consonants 8 Lateral Consonants 9 The Velar Nasal 10 Vowel Nasalization
S F O PRO
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
76
Mikael Parkvall 11 Front Rounded Vowels 12 Syllable Structure 13 Tone 14 Fixed Stress Locations 15 Weight-Sensitive Stress 16 Weight Factors in Weight-Sensitive Stress Systems 17 Rhythm Types 18 Absence of Common Consonants 19 Presence of Uncommon Consonants 20 Fusion of Selected Inflectional Formatives 21 Exponence of Selected Inflectional Formatives 22 Inflectional Synthesis of the Verb 23 Locus of Marking in the Clause 24 Locus of Marking in Possessive Noun Phrases 25 Locus of Marking: Whole-language Typology 26 Prefixing vs. Suffixing in Inflectional Morphology 27 Reduplication 28 Case Syncretism 29 Syncretism in Verbal Person/Number Marking 30 Number of Genders 31 Sex-based and Non-sex-based Gender Systems 32 Systems of Gender Assignment 33 Coding of Nominal Plurality 34 Occurrence of Nominal Plurality 35 Plurality in Independent Personal Pronouns 36 The Associative Plural 37 Definite Articles 38 Indefinite Articles 39 Inclusive/Exclusive Distinction in Independent Pronouns 40 Inclusive/Exclusive Distinction in Verbal Inflection 41 Distance Contrasts in Demonstratives 42 Pronominal and Adnominal Demonstratives 43 Third Person Pronouns and Demonstratives 44 Gender Distinctions in Independent Personal Pronouns 45 Politeness Distinctions in Pronouns 46 Indefinite Pronouns 47 Intensifiers and Reflexive Pronouns 48 Person Marking on Adpositions 49 Number of Cases 50 Asymmetrical Case-Marking 51 Position of Case Affixes 52 Comitatives and Instrumentals 53 Ordinal Numerals 54 Distributive Numerals 55 Numeral Classifiers 56 Conjunctions and Universal Quantifiers 57 Position of Pronominal Possessive Affixes
S F O PRO
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
How European is Esperanto?
58 Obligatory Possessive Inflection 59 Possessive Classification 60 Genitives, Adjectives and Relative Clauses 61 Adjectives without Nouns 62 Action Nominal Constructions 63 Noun Phrase Conjunction 64 Nominal and Verbal Conjunction 65 Perfective/Imperfective Aspect 66 The Past Tense 67 The Future Tense 68 The Perfect 69 Position of Tense-Aspect Affixes 70 The Morphological Imperative 71 The Prohibitive 72 Imperative-Hortative Systems 73 The Optative 74 Situational Possibility 75 Epistemic Possibility 76 Overlap between Situational and Epistemic Modal Marking 77 (Semantic Distinctions of Evidentiality) 78 (Coding of Evidentiality) 79 Suppletion According to Tense and Aspect 80 Verbal Number and Suppletion 81 Order of Subject, Object and Verb 82 Order of Subject and Verb 83 Order of Object and Verb 84 Order of Object, Oblique, and Verb 85 Order of Adposition and Noun Phrase 86 Order of Genitive and Noun 87 Order of Adjective and Noun 88 Order of Demonstrative and Noun 89 Order of Numeral and Noun 90 Order of Relative Clause and Noun 91 Order of Degree Word and Adjective 92 Position of Polar Question Particles 93 Position of Interrogative Phrases in Content Questions 94 Order of Adverbial Subordinator and Clause 95 Relationship between the Order of Object and Verb and the Order of Adposition and Noun Phrase 96 Relationship between the Order of Object and Verb and the Order of Relative Clause and Noun 97 Relationship between the Order of Object and Verb and the Order of Adjective and Noun 98 Alignment of Case Marking of Full Noun Phrases 99 Alignment of Case Marking of Pronouns 100 Alignment of Verbal Person Marking 101 Expression of Pronominal Subjects
S F O PRO
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
77
78
Mikael Parkvall 102 Verbal Person Marking 103 Third Person Zero of Verbal Person Marking 104 Order of Person Markers on the Verb 105 Ditransitive Constructions: The Verb ‘Give’ 106 Reciprocal Constructions 107 Passive Constructions 108 Antipassive Constructions 109 Applicative Constructions 110 Periphrastic Causative Constructions 111 Nonperiphrastic Causative Constructions 112 Negative Morphemes 113 Symmetric and Asymmetric Standard Negation 114 Subtypes of Asymmetric Standard Negation 115 Negative Indefinite Pronouns and Predicate Negation 116 Polar Questions 117 Predicative Possession 118 Predicative Adjectives 119 Nominal and Locational Predication 120 Zero Copula for Predicate Nominals 121 Comparative Constructions 122 Relativization on Subjects 123 Relativization on Obliques 124 ‘Want’ Complement Subjects 125 Purpose Clauses 126 ‘When’ Clauses 127 Reason Clauses 128 Utterance Complement Clauses 129 Hand and Arm 130 Finger and Hand 131 Numeral Bases 132 Number of Non-Derived Basic Colour Categories 133 Number of Basic Colour Categories 134 Green and Blue 135 Red and Yellow 136 M-T Pronouns 137 N-M Pronouns 138 Tea 139 (Irregular Negatives in Sign Languages) 140 (Question Particles in Sign Languages) 141 (Writing Systems) 142 Para-Linguistic Usages of Clicks
S F O PRO
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
How European is Esperanto?
79
Author’s address Institutionen för lingvistik Stockholms universitet SE-106 91 Stockholm, Sweden
[email protected]
About the author Mikael Parkvall is a linguist at Stockholm University. His previous work spans several areas of general linguistics, but has had a focus on pidgin and creole languages. Books published in English include Out of Africa (2000) and Limits of Language (2006).
S F O PRO
D E ANY T P M C O O R R E LISHING C
U ONH NCB E N J A M I N S P U B ©J
1st proofs
