Haitao Liu 37

Journal of Universal Language 7

September 2006, 37-64

Neutrality of International Languages*

Haitao Liu
Communication University of China

Abstract
This paper focuses on the neutrality of international languages. First,
a derivation of the concept of “neutral language” from “inter-
national communicative act” is provided; it is argued that an
acceptable neutral language for international communication can only
be an artificial language. Certain characterizations of consciously
created languages are discussed. The paper distinguishes two types of
neutrality: communicative neutrality and linguistic neutrality. All
planned languages are communicatively neutral, but their linguistic
neutrality varies, reflecting the diversity of language design
principles. Given that absolute linguistic neutrality unattainable, it
becomes reasonable to construct a language based on certain control
languages plus linguistic universals. We introduce the term
“deneutralization” to designate a process whereby a neutral language
changes into an ordinary language. The paper also shows that
Esperanto has not become deneutralized.

Keywords: international language, neutrality, deneutralization, artificial

language, planned language, universal language, Esperanto
*
I am grateful to Detlev Blanke for insightful discussion. I’d also like to thank
Probal Dasgupta for providing detailed comments and correcting my English.
38 Neutrality of International Languages

1. Introduction

Today English is already a global language (Crystal 1997), but

many people still consider that it is not an ideal solution for
international communication. Wright (2000: 246-247) mentions one
reason for this in her book on the role of language in nation-state
building and European integration: “They [artificial languages] are
ideally suited to the role [of lingua franca] since they are not the
languages of European nations of states. No section of the EU would
reject Esperanto in the way that it would reject English. None of the
objections would apply”. Gobbo (2005) also maintains that an
International Auxiliary Language (IAL) would be likely to serve as a
common language in the European Union more effectively than
English and other natural languages. If we designate this property of
artificial languages which Wright and Gobbo value highly—as the
neutrality of a language, it is appropriate to suggest that an ideal
means of international communication should be more neutral than
ethnic languages. But does any such thing as a neutral language
exist? Or are we to accept the claim of Van Parijs (2003) that “there
is no neutral language, no language equidistant from all others”?
How can we characterize more carefully the neutrality of a language
in international communication? In this paper, we will try to provide
an explicit characterization of the neutrality of a language in the
context of international communication. In the second section, we
discuss the general concepts of international language, universal
language, artificial language, planned language. The third section
focused on the notion of the neutrality of a language. The linguistic
neutrality of an international language is the topic of the fourth
section. The question of whether an artificial language would
deneutralize once the number of its native speakers increases is,
discussed in section 5.
 Haitao Liu 39

2. International Language and Artificial Language

Ammon (1994) provides a definition of international language

(henceforth IL): “a language used for international communication.”
By definition, international communication occurs between nations,
or rather, individuals belonging to different nations. A
communicative act may be called international only if it occurs
between different countries, i.e., between inhabitants or citizens of
different countries, or between different nationalities, i.e.,—roughly
speaking—between (native) speakers of different languages.
In this definition, we see no formal conditions structurally
constraining the class of ILs. In other words, any language can play
the role of an IL and be called an IL if it is used in an international
communicative act. The following figure makes explicit some points
about the roles a language can play in international communication
(Liu 2001a: 151):

La
A

Lc

Lb B

La, Lb and Lc are three (different) languages. La and Lb are the

mother tongues of communicative partners A and B respectively. It
is evident that A and B are communicatively equal in the interaction
only if Lc is not the same as either La or Lb. Accordingly, we can
call Lc a neutral language for A and B, because it is neither A’s nor
B’s mother tongue.
If we enlarge the set of communicators to include all the people
of the world, it is not possible to find a natural language that can
40 Neutrality of International Languages

serve as the neutral Lc. In international communication, if any

national language plays the IL role, the privileges of the native
speakers of the IL are immediately evident: (1) The native speakers
of the IL are not forced to study any foreign languages in order to
communicate internationally, but can spend their precious time on
other, perhaps more rewarding endeavors. (2) They are, as a rule,
linguistically superior in the IL to non-native speakers of IL, i.e.,
they can express their ideas more precisely, more grammatically,
more elegantly, and more fluently. That the native speaker’s person’s
privilege is the non-native speaker’s disprivilege felt quite acutely
by many communicators (Ammon 1994: 1729).
In a study on foreign language learning of European Union, Grin
(2005: 7) arrives at the following conclusions: (1) the United
Kingdom, in the EU context, gains at least 10 billion Euros per year
because of current predominance of English; (2) if one takes account
of the multiplier effect of certain components of this sum, which the
Anglophone countries can, because of the privileged position of
their language, invest elsewhere, this total becomes 17 to 18 billion
Euros per year; (3) this estimate does not take into account certain
symbolic system effects (such as the leverage that the native
speakers of a hegemonic language have in any situation of
negotiation or conflict in that language); it is clear that these effects
also have tangible economic consequences.
One response to this problem has been a series of efforts to
create a language specifically designed as a neutral medium for
international communication. Initiators of such projects designate
their object of construction variously as an international language,
an auxiliary language, an artificial language, a universal language, a
world language, or a planned language.
Thus, the term IL is used in two senses: (1) the common-sense
understanding is that a language that people from different
backgrounds or nations use with each other—IL as a function; (2)
the literature also conceptualizes ILs as entities, using the term ILs
 Haitao Liu 41

to refer to artificial or planned languages, languages that were

created specifically in order to facilitate international links and
understanding, sometimes termed international auxiliary languages
(Phillipson 1999: 24-25).
At the beginning of his monograph on artificial languages, Large
(1985) writes, “Since the early seventeenth century, several hundred
artificial language schemes have been constructed in the hope that a
universal medium for international communication can be adopted.
Unlike any natural language, which already possesses a group of
native speakers, an artificial language would represent, it is argued,
a neutral tongue acceptable to all [...] In some cases these language
schemes have been intended to act as a universal language in place
of all existing languages: one language for the world. More usually,
however, the language constructors have shouldered less ambitious
aims: to create an international auxiliary language which would
function for international communication alongside its parochial
natural cousins.”
This paragraph gives two basic features of such a construct: (a) a
universal medium for international communication; (b) neutral and
therefore acceptable to all. According to Large, artificial languages
can be classified into two subsets: universal languages and
international auxiliary languages. In my usage and that of other
scholars working on artificial languages, however, these two terms
are synonyms. Our usage goes back to the first classic
compendium-level study of artificial languages (Couturat & Leau
1903, 1979), whose authors explain that they use the expression
“langue universelle” (universal language) as a synonym of “langue
internationale auxiliaire” (international auxiliary language), because (1)
a “universal language” is not conceptually coterminous with the
notion of a (future) unique language of humankind; (2) modern
authors of “universal languages” do not expect their languages
suppress or supplant national languages.
The expression universal language was used primarily in the
42 Neutrality of International Languages

seventeenth, eighteenth, and nineteenth centuries, but in our era, the

term “universal language” can also be found in the titles of works on
international language or in the names of projects. A simple word
count of the relevant occurrences in Duličenko (1990) shows that
the names of more than 70 artificial language projects contain the
word “universal” or its translation equivalents. Several major
scholars in interlinguistics use the term “universal language” in the
titles of their works (Knowlson 1975; Couturat & Leau 1903, 1979;
Slaughter 1982; Strasser 1988). Today, we should also add the name
of Journal of Universal Language to this list.
According to Duličenko (1989: 54), a typical universal language
(lingua universalis, langue universelle, vseobščij jazyk, and so on)
was designed for use by all people on Earth and forever (i.e., as the
foremost or unique human language of the future). Projects of this
kind were actively worked out from the 17th century (the first project
bearing the name of “universal language” appears in 1650) until the
middle of the 19th century. In the second half of the 19th century, the
popularity of “universal language” projects declined, and in the 20th
century they occur only sporadically.
Eco proposes a distinction between a perfect language and a
universal language. He distinguishes between the search for a
language capable of mirroring the true nature of objects—a perfect
language—and the search for the language which everyone might,
or ought to, speak—a universal language (1995: 73). It seems to me
that Eco’s perfect language corresponds to my understanding of
universal language, while his universal language is the familiar
notion of an auxiliary language for international communication.
Therefore, the term “universal language” has two basic meanings:
(1) the only language of a future united humankind, described
occasionally in utopian social models in the Renaissance and in
more recent works of utopian socialism or modern interlinguistics;
(2) a consciously created language for international communications.
Blanke (1997: 5) sees a mismatch between the second meaning and
 Haitao Liu 43

the term “universal”, which he thinks gives the erroneous

impression that the purpose of such a language is to push aside and
replace national languages. This is indeed a possible
misunderstanding. Some of my students ask me if the aim of
shijieyu (the Chinese name for Esperanto) is to replace the national
language. More generally, the term universal language tends to give
rise to negative ideas about the function of such a language.
If the term universal language is not a suitable name, why is it in
use? Künzli (2006) explains the preference for the term “neutral
universal language” (neutrale Universalsprachen) over “international
planned languages” (internationale Plansprachen). He believes that
the former is easier for layman to understand and also less
misleading than the latter, which often also include other systems
under the term planned language.1 In principle, then, the two terms
are the same. Another factor is the fact that some constructors call
their system ‘universal’ because the system is based on linguistic
universals.
Blanke (1989: 63) analyzes the various designations for the
artificial languages and recommends the term “planned languages”,
defined as “language systems which have been consciously created
according to definite criteria of an individual or group of individuals
for the purpose of making international communication easier”.
The definitions given above can be schematically represented as
follows:

1
For example, programming languages for computer, and interlanguages for
contrastive linguistics.
44 Neutrality of International Languages

Natural International Languages

International Languages
International Auxiliary
Languages or International
Planned Languages
Planned Languages Universal Languages
Artificial Languages
Programming Languages for Computers

In the following sections, in accordance with what Blanke

recommends, we use the term “planned language” to designate a
language consciously created for international communication, and
we regard this term as synonymous to “international auxiliary
language” and “international planned language”. These terms are
used in the titles of two important works Blanke (1985) and
Duličenko (1990).

3. The Neutrality of an International Language

From the discussion above, it is clear that any language can play
a role of neutral international language, but a natural language
playing the IL role is limited to regional functioning: it cannot serve
as a global neutral language. A global neutral language can only be a
planned language, for only such a language is neither the mother
tongue of any ethnicity nor the national language of any nation or
state. If this plausible line of reasoning does indeed hold, why does
Van Parijs (2003) say, “There is no neutral language”? To clarify this
remark, he explains that Esperanto, for example, belongs
unambiguously to the Western group of Indo-European languages,
with identifiable Romance and Germanic ingredients. On that matter,
he has a point. At least lexically speaking, Esperanto is a Romance
language. Gledhill (2000) analyzes the language sources of the most
 Haitao Liu 45

often used 100 and 1,000 words in Esperanto. His findings are as
follows (“Esperanto” is his term for words formed by internal
combinatorial devices peculiar to Esperanto itself, and “Indo-
European” is to be construed as meaning “other Indo-European”.):

Romance Esperanto German Indo-European Greek Balto-Slavic

70% 12% 10% 5% <2% <1%

If the planned language can be linguistically shown to belong to

a historically specifiable family of natural language, it would not be
neutral for learners from other languages families. This line of
reasoning underwrites the statement by Van Parijs (2003) that “when
proposed on a world scale, or even within Europe with Finnish,
Estonian, Hungarian, Basque and Maltese as part of the picture, it
[Esperanto] cannot make any claim to neutrality.”
It is clear that what Van Parijs means by neutrality is quite
different from the notion of neutrality elaborated in section 2.
In order to articulate the neutrality notion further, beyond these
preliminaries, it is useful to review the definitions of neutrality
provided by other interlinguists.
Kuznecov (1991: 206) defines neutrality as “the distinguishing
feature of an artificial language with respect to the natural (national)
languages, to be understood as independence from political,
economic and other interests of this or that nation. Dead ethnic
languages also have such neutrality—for example, Latin, when used
as an international language. Neutrality can also be interpreted as an
intrinsic property of an artificial language, if its structure has been
designed so as to explicit not resemble the structure of any of the
source languages.”
Szilágyi (1931: 76) defines the neutrality-balance of the
international language as “a feature of the international language.
46 Neutrality of International Languages

The principle of neutrality-balance requires that the international

language be equidistant from all national languages. If these
languages appear as constituents within an international language,
then they should be evenly distributed.”
Schubert (2004: 328) introduces a new concept of “intercultural
language”. Every language has an inherent “interculturality”, he
argues, which determines the potential of that language as a means
for intercultural communication. We may regard this interculturality
as a notion closely related to that of the neutrality of language. The
following figure is adapted from Schubert.

Controlled Languages
Pidgins

Lingua Franca

Ethnic Languages Planned Languages

Low Degrees of Interculturality High

The figure displays clearly the advantage of a planned language

as a neutral medium of intercultural communication. It is also worth
remarking that Pidgins also have a high interculturality (or neutrality)
compared with ethnic languages.
On the basis of this survey of the positions articulated in the
literature, it seems reasonable to distinguish two major aspects of
2
neutrality : (1) communicative neutrality, or the requirement that the
international language should not be the mother tongue of any
participant in the communicative act; (2) linguistic neutrality, or the
requirement that the international language should be linguistically
equidistanct from the participant’s mother tongues, and thus from all

2
These two aspects are similar with the classification of Detlev Blanke (2006,
personal communication): political neutrality and linguistic neutrality.
 Haitao Liu 47

languages of the world. While the planned language seems an

adequate solution to the problem of communicative neutrality, the
linguistic neutrality of a planned language is not a problem that is
easy to address.
Bastardas i Boada (2002) too drawn a similar conclusion:
“Clearly, if a neutral code that is not the L1 of any group was
adopted, people would be less likely to see a code of
intercommunication as an L1, thus guaranteeing further the level of
conservation of historical linguistic diversity. This would also make
humans more equal in terms of their initial language competencies,
since everybody would have to learn the language. However, here
we may face problems such as the linguistic distance between the
languages of each group and the structure of the language of
intercommunication that is finally adopted. How can we create a
neutral code that will be equal for everybody?”

4. The Linguistic Neutrality of a Universal

Language

According to the traditional classification by Couturat & Leau

(1903, 1979), which is based on the relationship of planned
languages to ethnic languages, planned languages can be categorized
as a-priori or a-posteriori or mixed systems. An a-priori language is
composed entirely of invented elements not found in any natural
language, and is usually based on a logical classification of ideas. In
practice, this tends to make the language more difficult to learn and
use. An a-posteriori language is based on elements of grammar,
vocabulary, and syntax drawn from one or more natural languages.
This more pragmatic approach is usually motivated by a desire to
design a workable auxiliary language, which can be easily learned
and used by everyone. A mixed system includes both elements. This
classification is evidently related to the lexical material of planned
48 Neutrality of International Languages

languages. In other words, we can place any system of planned

language on an axis ranging from artificiality to naturalness. In fact,
we can even apply these criteria to natural languages as well. This
involves highlighting the fact that natural languages have unnatural
(artificial) elements just as planned (artificial) languages have
natural elements. Schubert (1989) constructs a continuum of
decreasing artificiality that ranks the following types of languages:

- An a-priori planned language (e.g., Leibniz’s language of 1666)

- An autonomous a-posteriori language (Esperanto)
- A naturalistic a-posteriori language (Occidental)
- A compromise language for a certain family of ethnic languages
(pan-Slavic)
- A modified or simplified ethnic language (Latine sine flexione)
- A simplified ethnic language for the purpose of introducing
learners to the unmodified language (Basic English)
- A more or less consciously developed literary language linking
several language communities (medieval Franco-Italian)
- A highly planned ethnic language (Estonian)
- A super-regional standard form of an ethnic language (High
German)
- An ethnic language “restored” by purists (Icelandic)
- An “untouched” ethnic language (Frisian)

This viewpoint is not only accepted by interlinguists; we even

find a similar language spectrum from an article about computer
languages (Baron 1994: 663):

NATURE
- Animal signaling systems
- Natural human languages
- Sublanguages
- Universal languages based on natural languages
 Haitao Liu 49

- Logically constructed universal languages

- Computer languages
- Notation schemes
ARTIFACT

If a planned language is preferred merely for reasons of

“neutrality”, then it seems obvious that the choice should fall on an
a-priori language, having no connection with any known tongue, but
rather serving as a brand-new vehicle of human thought (Pei 1958:
174). But the central function of an international auxiliary language
is to serve as the medium of international communication. In our
quest for a perfect and completely neutral language, we ought not to
ignore the history and practice of planned languages.
Blanke (1985) establishes 18 stages of planned language
development from a project to a language, on the basis of practical
usage. 3 We emphasize the importance of the sociolinguistic
classification of Blanke to show that social factors are truly
influential or decisive in the development of planned languages. The
serious study of planned languages cannot afford to ignore the social
factors. According to the practical criteria of Blanke, we can classify
approximately 1,000 planned language systems into three classes:
PL (planned language) projects, semi-PLs, and PLs. Almost all the
systems belong to the project category. The exceptions include a few
systems that reached stages 15-16, even 19 (of the 28 stages). They
are Volapük (Schleyer 1879), Latino sine flexione (Peano 1903), Ido
(Couturat 1907), Occidental (Edgar de Wahl 1922), Basic English
(Ogden 1930), and Interlingua (IALA/Gode 1951)4; these count as
semi-PLs. The exception among these exceptions, Esperanto
(Zamenhof 1887) has to some extent crossed all 28 thresholds and

3
He later breaks the process down into 28 stages (Blanke 2001).
4
The parenthesized indications are not to be construed as bibliographic references;
what we show is the name of the author and the year of inception of the project.
50 Neutrality of International Languages

become the only PL. All the semi-PLs Blanke considers are based
on natural languages and can be classified as a-posteriori or mixed
systems. The history of planned languages shows that the task of
interlinguists is to find a balancing point between a-priori and
a-posteriori.
Among the principles for constructing planned languages, the
first principle is “Base any arbitrary properties in the planned
language on the corresponding properties in the languages of the
target population. Aim to represent all subgroups equally in this
respect.” But, just said by the author “if we accept the premise that
languages of the target population should in some ways influence
the properties of the language that is to be used as a vehicle of
communication between them, then it is clear that no one language
can be the ideal planned language for all of these different
populations, since different target populations will have languages
with different linguistic properties, and these should be reflected in
the corresponding target languages” (Maxwell 1989: 103-104).
In 1908 Otto Jespersen, in his preface to the Ido-German
dictionary by Louis de Beaufront, formulated, in Ido, an important
facility maximization principle for interlinguistics: ‘that international
language is best which in every point offers the greatest facility to
the greatest number’ (la maxim bona internaciona linguo helpanta
esas ta, qua en omna punti ofras la maxim granda facileso a la
maxim granda nombro de homi). In his book “An International
Language”, he enlarged upon that principle by saying that it does
not mean that we should take Chinese as our interlanguage, simply
on the basis of the fact that it is known to the greatest number of
men. In other words, the principle does not apply, as he made clear
on several occasions, to an absolute number of individuals, but only
to the number of those individuals who require communication with
other nations.
Jespersen (1928) also argues “It is, however, very important to
remember that the facility of which we speak here is not merely the
 Haitao Liu 51

superficial facility, with which a printed message can be understood

at first sight—that is something, but not everything. For an
interlanguage to be really useful it must be easy not only to the
reader, but also to the intending writer and speaker, and this implies
a good deal more”. Yes, facility maximization, rigorously understood,
is a very important principle, which tells us that the selection of
lexical material is only one aspect of making a language easy to
learn; a language designer also needs to consider the other
components of a language. Linguistic neutrality cannot involve only
lexical neutrality, but must also include other aspects of linguistic
structure.
Jespersen maintains that the creators of an international language
need only consider the languages of those people who require
international communication. Janton (1988: 1681) revisits this point:
“An IL (international language) is not only an interethnical or even
international language since it can be non-territorial and non-ethnic.
In order to define the internationality of a language, both the
quantity and the quality of communication realized in this language
must be taken into consideration. That is why it is more important
for the expansion of a language to conquer a sphere of activity than
a large population with a small amount of communication.” So
understood, neutrality seems to amount to no more than the
internationality of the lexicon of a language. It is from this
viewpoint that Ido has been projected as having a more international
appearance to it than Esperanto. Here is a more precise analysis of
the Ido lexicon by Couturat: 91 % of the words match French, 83 %
match Italian, 79 % match Spanish, 61 % match German, 52 %
match Russian. These estimates, cited in Jacob (1947: 93), do not
add up to 100, since the same Ido word can find several matches. It
is obvious that Ido is distinctly more Romance (or Latinate) in its
vocabulary than Esperanto. This can look like an advantage only
because, in the eyes of many creators of planned languages, a
Latinate visage may suffice for “international” branding.
52 Neutrality of International Languages

Barandovská-Frank (1995: 97-101) provides a list of Latin-based or

Latinate planned language systems; her list includes 179 projects
until 1993.
Critical readers are of course bound to ask: if creators of planned
languages are aiming at international application, why do they prefer
a Romance lexicon? Why can’t we make a more international
lexicon on the basis of a mathematical and statistical calculation?
Concerning the last point, Janton (1993: 137-138) has a good
explanation: “It would be naive to suppose that internationality
consists primarily in the greatest diversity of lexical sources. A
language comprised of words from all languages would be
statistically but not linguistically international. The proportion of
each language in the whole would be so small that the sum of such
contributions would seem foreign to everyone. It has been calculated
that we can understand 80 percent of a language by means of only
two thousand words. If we were to choose, say, one hundred
languages (thereby eliminating 96 percent of the spoken languages
of the world), the share of any one would be twenty words. What
would be the advantage of that?”
The idea of a statistically equitable neutral IL has in fact been
implemented; we need not take Janton’s speculations at face value.
Brown (1960) tries mathematically to construct the words of his
system “Loglan”. His paper offers the following account of how the
word-shapes were arrived at. According to his calculation, over two
thirds of the world’s present inhabitants speak one or more of
exactly eight of its several hundred natural languages with either
native or second language proficiency. Counting, for each language,
both its native speakers and those of its proficient non-native
speakers who are not native speakers of any of the other seven, these
eight languages, in the approximate descending order of the number
of their proficient speakers, are: English, Mandarin Chinese, Hindi,
Russian, Spanish, Japanese, French and German. Now if one regards
the 1700 million speakers of the eight major languages as the target
 Haitao Liu 53

population of Loglan research, the relative statistical importance of

each of them may be defined as the proportion of their speakers in
the whole. On that basis the relative importance of English is
approximately 0.28; Chinese 0.25; Hindi 0.11; Russian 0.10; and so
on down through German, with 0.05. If these figures are even
approximately correct, English and Chinese are overwhelmingly the
most “important” modern languages; their speakers constitute 53 per
cent of the target population. We turn now to Brown’s methodology
for word shape choice. The following is an example of choosing a
shape for word for “blue”, which comes out as “BLANU” in Loglan:

BLANU ALL OF ENGLISH BLUE [BLU] 1 X 0.28 =0.28

BLANU ALL OF CHINESE LAN 1 X 0.25=0.25
BLANU 1/2 OF HINDI NILA 0.5 X 0.11=0.06
BLANU 2/7 OF RUSSIAN GALUBOI 0.3 X 0.10=0.03
BLANU 1/2 OF SPANISH AZUL [ASUL] 0.5 X 0.09=0.05
BLANU NO COUNTABLE PORTION OF 0 X 0.06=0.00
JAPANESE
AO OR KON
BLANU 2/3 OF FRENCH BLEU [BLÖ] 0.7 X 0.06=0.04
BLANU ALL OF GERMAN BLAU 1 X 0.05=0.05
TOTAL LEARNABILITY SCORE =0.76

Only the phonemes common to and occurring in the same order

in both the Loglan and the natural language word are counted
towards determining the learnability score. The total learnability
score expresses the probability that a person will learn the word
from association with a familiar word in his base language.
The process is mathematically precise and scientific, but the
outcome is less acceptable than Brown might expect. As a native
speaker of Chinese it is not an easy task for me to recognize
BLANU as Chinese LAN. For a native speaker of Hindi, the fact
54 Neutrality of International Languages

that the phonemes LA appear in the Hindi word for ‘blue’, NILA is
less salient than the fact that the same phonemes LA appear in the
Hindi word for ‘red’, LAL!5
Thus, Blanke (1985: 95) is right about the pointlessness of trying
to ensure absolute internationality in a planned language: “A kind of
absolute internationality would be reached, if in the vocabulary of a
planned language all language of the world (proportional to its
number of speakers) were represented. Such internationality would
not be useful to anybody. The vocabulary would be extraordinarily
heterogeneous and would be helpful for nobody.”
Given that we cannot construct a viable language representing all
linguistic properties drawn from the whole world on a meaningful
basis, it is a rational decision to select some languages as one’s
control languages.
If the matter is put like this, we have to consider how to choose
these control languages. It is plausible to propose the natural
languages that play the IL role as suitable candidates.
Ammon defines “the international standing of a language” as
“the extent to which the language is actually used for international
communication, i.e., for communication between different nations”
(2003: 231). There are some indicators for determining an
international standing of a language: numerical strength, economic
strength, political strength, cultural strength and pedagogic strength.
These indicators are also useful in the context of the choice of
control languages when constructing or evaluating a planned
language.
According to the formula for calculating the communication
potential (Q) of a language mentioned by De Swaan (2001: 33), the
Q-value of a language is the product of two numbers: P and C. Here
P is the prevalence of the language; it stands for the percentage of
the total world population that speaks it fluently. C is the value of

5
Thanks to Probal Dasgupta for providing this interesting example.
 Haitao Liu 55

the language as a common interlanguage; it stands for the

percentage of the speakers of the language who have learnt more
than one language. It is a generally appreciated fact that bi- and
multilingual speakers connect the multilinguistic world into a whole.
This being the case, Q can be taken to express the connecting
capacity of the language. Thus, it is also acceptable to select some
high Q-value languages as control languages when one proposes to
construct or evaluate a planned language.
In summary, it is clear that linguistic neutrality is not an absolute
concept. A language built on the basis of some purely formal
absolute neutrality principle would not work as a language for
humankind, because it would also have to fall in line with the
6
known universals in human languages. It would appear to be
rational to create or evaluate a planned language based on (a) some
control languages selected on the basis of the international standing
of the relevant languages and (b) linguistic universals. Such a
procedure leads to a system that is a mixed language, with distinct
internationality profiles on different planes of linguistic structure.
For instance, lexically, Esperanto can be considered mainly a
Romance language. Morphologically, it is an agglutinating language
with a strong similarity to isolating languages. At the levels of
syntax and style, it exhibits a significant degree of Slavic influence.
Functionally, it has served as an interlanguage for more than a
century (Janton 1973, 1993; Piron 1981; Wells 1989). Nuessel gives
Esperanto the following properties: “a planned, a posteriori
language, an amalgamation of the linguistic elements of the various
ethnic languages including Yiddish, Germanic, and Slavic tongues
that were a part of Zamenhof’s socially rancorous environment. The
language also contained grammatical features of certain Romance
languages with which Zamenhof was familiar” (Nuessel 2000: 41).
Unish, a system developed by the research team at Sejong

6
Liu (2006) presents an absolutely neutral planned language project.
56 Neutrality of International Languages

University, is also an interesting and significant new planned

language.7 Its lexical material is drawn from 15 languages (English,
Spanish, Portuguese, Italian, French, German, Russian, Korean,
Chinese, Japanese, Arabic, Hindi, Greek, Latin, and Esperanto). The
basic principles for developing this new language are “ease” and
“commonality” (Kwak 2003). The selection of control languages
and other aspects of the methodology of its construction lead us to
believe that the creators of Unish are trying to build an international
language as linguistically neutral as possible. A comparison between
Unish and Esperanto can be found in Lee (2002).

5. Deneutralization of an International
Planned Language

There is more to say about the neutral language notion, however,

because “even if some world-wide neutral language had been found,
nothing would prevent it, after some generations, from thickening
from a lingua franca into the mother tongue of some, as happened to
Swahili, for example, with the consequence that once again
neutrality would be lost and the whole process of designing a neutral
language in need of being relaunched” (Van Parijs 2003). We call
the process that a language changed from neutral to ordinary status
as “deneutralization” of language.
We can compare the process of planned language development
with the creolization of pidgins.8 When a pidgin has enough native
speakers, it is creolizated. In our term, it is also deneutralized.
As mentioned above, Blanke establishes 18 stages of a planned

7
Unish means a universal language targeted to a lingua franca in the international
communication. http://www.unish.org/.
8
A detailed comparison between planned languages and pidgins/creoles is carried
out in Liu (2001b). Kwak (2003) provides a comparison between Unish and
pidgins.
 Haitao Liu 57

language development from a project to a language, on the basis of

practical usage. Of these, the last stage is the appearance of a
bilingual speaker who has learned the planned language from birth
(Blanke 1985: 112). Thus, the emergence of native speakers is an
important milestone in the evolution of a planned language. In the
following discussion, we are able to use Esperanto as an example.
The absence of native speaker data in the case of other PLs or
semi-PLs makes it impossible to attempt any meaningful
comparisons in this domain.
It is possible, though, to find parallels in the case of the evolution
of creoles. It is well known that native speakers are just as important
or more in the development of a pidgin language; if a pidgin
language is acquired by sufficiently many native speakers, then the
pidgin will become a creole (though its speakers may choose not to
call it that). For example, Mühlhäusler considers the process of
creolization to be divisible into the following stages: “Jargon
(pre-pidgin, multilingual idiolect, secondary hybrid)—Stable Pidgin
(pidgin, basilectal pidgin, tertiary hybrid)—Expanded Pidgin
—Creole” (Mühlhäusler 1997: 6).
Perhaps, in view of the importance of native speakers in the
development of the language, Esperanto should be considered a
creole language or at least some kind of object of “creolizatino”
studies? Does creolization really take place in Esperanto? For
serious creolization to occur, it is constitutively necessary that there
be enough native speakers who use the language in question as their
everyday first language. According to Corsetti (1999), the
percentage of native speakers in Esperanto is only 4%, far from 10%
of the total number—the proportion minimally required for
creolization from pidgins. Not only are the native speakers in
Esperanto not numerous enough; furthermore, the community is a
diaspora. Native speakers hardly use Esperanto on a daily basis
outside their families. In their case, Esperanto is only used in the
family domain. Essentially, Esperanto is still their second or third
58 Neutrality of International Languages

language. We can thus say, with Corsetti (1999: 47), that “from this
viewpoint Esperanto is not in a state to be considered a creole
language by creolists”. Schubert also reaches a similar conclusion:
“What about creolization in planned languages? At least for
Esperanto, there are indeed persons who speak it as their native
language. But their number, possibly a few hundred, is small
compared with the language community, and they have no special
standardizing influence of the development of Esperanto […] The
language community as a whole is a pure second-language
community. Esperanto, the planned language that has grown farthest
into communicative use, is far from creolization” (Schubert 1989:
11). Although Esperanto functions mainly as a second language,
nevertheless, it is not similar to other languages having this function.
Wood rightly says, “the status of Esperanto as a second language for
its users is different from that of ethnic languages which have been
acquired as second languages by learners faced by the economic,
political and educational pressures which make such acquisition
necessary or desirable” (Wood 1979: 435).
Versteegh (1993: 593) argues “the acquisition of Esperanto as a
first language is a special case of language acquisition with
restricted input, since the monitoring parents are not native speakers
of Esperanto themselves. Consequently, the case of the denaskaj
esperantistoj (native speaker esperantists) may be compared with the
process of creolization, in which children acquire a language variety
that is by definition not the native language of the parents”. If we
follow Versteegh’s definition in understanding the notion
“creolization”, creolization of Esperanto is comparable to
creolization of pidgins. Therefore, we think that Gledhill (2000: 42)
is also right: “The relationship between Esperanto and creoles is
therefore more abstract: the transformation of Pidgins into Creoles
may be reflected in the development of Esperanto from a schematic
design to a relatively widely-used language”. It follows that the
“creolization” of pidgins is somewhat similar to the socialization of
 Haitao Liu 59

planned languages.
Moreover, as we saw in Ferguson’s diglossias, complementary
distribution contributes to maintenance: the formal variety is not
habitually used in everyday communication and therefore rarely
becomes a mother tongue (Bastardas i Boada 2002).
In short, the statements about the creolization of Esperanto that
occasionally surface in the literature are only metaphorical. As far as
its status is concerned, a planned language is more similar to a
pidgin than to a creole, although it is intended for wider usage than a
pidgin is. In other words, indicators of deneutralization (in the sense
derived from the work of Van Parijs) have not yet become
observable even in the most developed planned language Esperanto,
where one would a prior imagine that the conditions for
deneutralization would be most favorable. The auxiliary function of
a planned language, we conclude, keeps it neutral for a very long
time indeed. The conclusions Van Parijs seems to have drawn from
such cases as Swahili possibly have to do with the fact that
Esperanto is indeed a planned language, whereas neither Swahili,
nor Indonesian, nor Modern Hebrew is really in the same league as
Esperanto as far as social realities are concerned.

6. Conclusion

At the beginning of the paper, we offered a working definition of

the notion of an international language. The conceptualization of an
international communicative act leads, in this paper, to the idea of a
neutral language. A neutral language for international communi-
cation can only be a planned language. Certain characterizations of
the consciously created languages are discussed, leading to the
choice of the least controversial term “(international) planned
language” for our purposes. In order to articulate adequately the
notion of neutrality, we focus on the neutrality of an international
60 Neutrality of International Languages

auxiliary language or international planned language in terms of a

distinction between two types: communicative neutrality and
linguistic neutrality. All planned languages are communicatively
neutral, but their linguistic neutrality varies, reflecting the diversity
of language design principles. Communicative neutrality involves
all users having to learn the language in order to be able to use it as
a means of communication; linguistic neutrality has to do with
maximizing equality of access for the learners with different mother
tongues.
Evidently, it is not an easy task to construct a language
linguistically equidistant from all the languages of the world. In
practice, absolute linguistic neutrality is neither practicable nor a fair
representation of our task, because our goal is to create a language
for humankind, which involves taking language universals on board.
In this perspective, constructing a language based on some control
languages coupled with systematic attention to linguistic universals
is perhaps a rational procedure, if the control languages are selected
from the set of languages that are in international use.
We also introduced a concept of “deneutralization” to express a
process whereby a neutral language changes into an ordinary
language as a consequence of the proportion of its native speakers
rising beyond a critical threshold. We consider the case of Esperanto,
inquiring whether the relevant processes in Esperanto are
comparable to the creolization of pidgins, a better understood
phenomenon. Our finding is that there have been no indicators of
deneutralization in the planned language in the 119 years of its
existence.

References

Ammon, U. 1994. International Languages. In R. Ashier (ed.), The

Encyclopedia of Language and Linguistics 4, 1723-1730. Oxford:
 Haitao Liu 61

Pergamon Press.
___________. 2003. The International Standing of the German Language.
In J. Maurais & M. Morris (eds.), Languages in a Globalising World
231-249. Cambridge: Cambridge University Press.
Barandovská-Frank, V. 1995. La Latina kiel Interlingvo. Dobřichovice:
Kava-Pech.
Baron, N. 1994. Computer Languages. In R. Asher (ed.), The Encyclopedia
of Language and Linguistics 2, 663-670. Oxford: Pergamon Press.
Bastardas i Boada, A. 2002. World Language Policy in the Era of
Globalization: Diversity and Intercommunication from the Perspective
of ‘Complexity’. Available at http://cultura.gencat.net/ llengcat/noves.
Blanke, D. 1985. Internationale Plansprachen. Berlin: Akademie- Verlag.
_________. 1989. Planned Languages: A Survey of Some of the Main
Problems. In K. Schubert & D. Maxwell (eds.), Interlinguistics:
Aspects of the Science of Planned Languages. Trends in Linguistics 42,
63-87. Berlin: Mouton de Gruyter.
_________. 1997. The Term “Planned Language”. In H. Tonkin (ed.),
Esperanto, Interlinguistics, and Planned Language 1-20. Lanham, New
York, & Oxford: University Press of America.
__________. 2001. Vom Entwurf zur Sprache. In K. Schubert (ed.),
Planned Languages: From Concept to Reality 37-89. Brussels:
Hogeschool voor Wetenschap en Kunst.
Brown, J. 1960. Loglan. Scientific American 202, 53-63.
Corsetti, R. 1999. Kreoliĝo de Esperanto: Inter Personaj Gustoj kaj Oficiala
Normo. In D. Blanke, R. McCoy, & O. Buller (eds.), Por Aktiva
Lingvopolitiko 43-60. Rotterdam: Universala Esperanto- Asocio.
Couturat, L. & L. Leau. 1903. Histoire de la langue universelle.
Hildesheim, Zürich, & New York: Olms.
___________________. 1979. Les nouvelles langues internationales.
Hildesheim, Zürich, & New York: Olms.
Crystal, D. 1997. English as a Global Language. Cambridge: Cambridge
University Press.
De Swaan, A. 2001. Words of the World. The Global Language System.
62 Neutrality of International Languages

Cambridge: Polity Press.

Duličenko, A. 1989. Ethnic Language and Planned Language. In K.
Schubert & D. Maxwell (eds.), Interlinguistics: Aspects of the Science
of Planned Languages. Trends in Linguistics 42, 47-61. Berlin: Mouton
de Gruyter.
Duličenko, А. 1990. Mezdunarodnye Vspomogatel'nye Jazyki. Tallinn:
Valgus.
Eco, U. 1995. The Search for the Perfect Language. Oxford: Blackwell.
Gledhill, C. 2000. The Grammar of Esperanto: A Corpus-based
Description. München: Lincom Europa.
Gobbo, F. 2005. The European Union’s Need for an International Auxiliary
Language. Journal of Universal Language 6, 1-28.
Grin, F. 2005. L'enseignement des langues étrangères comme politique
publique. Available at http://cisad.adc.education.fr/hcee/documents/
rapport_Grin.pdf.
Jacob, H. 1947. A Planned Auxiliary Language. London: Dobson.
Janton, P. 1988. Plans for an International Language. In U. Ammon, N.
Dittmar, & K. Mattheier (eds.), Sociolinguistics 1679-1687. Berlin:
Walter de Gruyter.
Janton, P. 1993. Esperanto: Language, Literature, and Community. Albany,
NY: State University of New York Press.
Jespersen, O. 1928. An International Language. London: Allen & Unwin.
Knowlson, J. 1975. Universal Language Schemes in England and France
1600-1800. Toronto & Buffalo, NY: University of Toronto Press.
Künzli, A. 2006. Neutrale Universalsprachen als Alternative: Einige
Pragmatische Überlegungen und Anmerkungen zum Thema. In A.
Künzli (ed.), Universalaj Lingvoj en Svislando. Svisa Enciklopedio
Planlingva 41-68. La Chaux-de-Fonds: CDELI/SES.
Kwak, E. 2003. Comparision between Pidgins and ‘Unish’. Journal of
Universal Language 4, 17-31.
Kuznecov, S. 1991. Kratkiy Slovar Interlingvistiki Terminov. In M. Isaev
(ed.), Problemy Mezdunarodnogo Vspomogatel'nogo Jazyka 71-228.
Moskva: Nauka.
 Haitao Liu 63

Large, A. 1985. The Artificial Language Movement. Oxford: Blackwell.

Lee, D. 2002. A Comparison of Unish Grammar with Esperanto. Journal of
Universal Language 3, 57-74.
Liu, H. 2001a. Informadika Aspekto de Interlingvistiko. In S. Fiedler & H.
Liu (eds.), Interlingvistikaj Studoj: Festschrift für Detlev Blanke zum 60
Geburtstag 147-171. Dobřichovice: Kava-Pech.
______. 2001b. Pidgins, Creoles, and Planned Languages: Linguistic
Development under Special Conditions. In K. Schubert (ed.), Planned
Languages: From Concept to Reality 121-177. Brussels: Hogeschool
voor Wetenschap en Kunst.
______. 2006. Kio Estas KOD? Kiel ĝi Funkciu? Informilo por
Interlingvistoj 2, 2-5.
Maxwell, D. 1989. Principles for Constructing Planned Languages. In K.
Schubert & D. Maxwell (eds.), Interlinguistics: Aspects of the Science
of Planned Languages. Trends in Linguistic 42, 101-119. Berlin:
Mouton de Gruyter.
Mühlhäusler, P. 1997. Pidgin and Creole Linguistics. London: University
of Westminister Press.
Nuessel, F. 2000. The Esperanto Language. New York, Ottawa, & Toronto:
Legas.
Pei, M. 1958. One Language for the World. New York: Biblo and Tannen.
Phillipson, R. 1999. International Languages and International Human
Rights. In M. Kontra, R. Phillipson, T. Skutnabb-Kangas, & T. Varady
(eds.), Language: A Right and a Resource. Approaching Linguistic
Human Rights 25-46. Budapest: Central European University Press.
Piron, C. 1981. Esperanto: European or Asiatic Language? Rotterdam:
Universala Esperanto-Asocio.
Schubert, K. 1989. Interlinguistics: Its Aims, its Achievements, and its
Place in Language Science. In K. Schubert & D. Maxwell (eds.),
Interlinguistics: Aspects of the Science of Planned Languages. Trends
in Linguistics 42, 7-44. Berlin: Mouton de Gruyter.
__________. 2004. Interkulturelle Sprache. In I. Müller (ed.), Und sie
Bewegt sich Doch: Translationswissenschaft in Ost und West.
64 Neutrality of International Languages

Festschrift für Heidemarie Salevsky zum 60 Geburtstag 319-331.

Frankfurt am Main u.a.: Lang.
Slaughter, M. 1982. Universal Languages and Scientific Taxonomy in the
Seventeenth Century. Cambridge: Cambridge University Press.
Strasser, G. 1988. Lingua Universalis. Kryptologie und Theorie der
Universalsprachen im 16. und 17. Jahrhundert. Wiesbaden:
Harrassowitz.
Szilágyi, D. 1931. Versus Interlinguistica. In Haupenthal (ed.),
Plansprachen. Beiträge zur Interlinguistik 163-187. Darmstadt: Wiss.
Buchgesellschaft.
Van Parijs, P. 2003. Europe’s Three Language Problems. Prepared for D.
Castiglione & C. Longman (eds.), The Challenge of Multilingualism in
Law and Politics. Oxford: Hart. Available at <http://www.etes.ucl.
ac.be/Publications/dochs.htm>.
Versteegh, K. 1993. Esperanto as a First Language: Language Acquisition
with a Restricted Input. Linguistics 31, 539-555.
Wells, J. 1978. Lingvistikaj Aspektoj de Esperanto. Rotterdam: Universala
Esperanto-Asocio.
Wood, R. 1979. A Voluntary Non-ethnic, Non-territorial Speech
Community. In W. Mackey & J. Ornstein (eds.), Socio-linguistic
Studies in Language Contact 433-450. New York: The Hague & Paris:
Mouton.
Wright, S. 2000. Community and Communication: The Role of Language in
Nation State Building and European Integration. Clevedon: Multilingual
Matters.