语言与人口：计划生育政策及人口变化对语言和语言学的启示

李嵬

自2005年以来，中国政府发布了一系列中国社会生活中的语言使用状况报告。这是中国在语言研究方面的重大进展，也引起了海外学界的极大关注。德国的德古意特出版社将这些报告中的一些文章译成英文，并结集出版。这些报告重新将我们的注意力聚焦在社会政策与社会变化、语言使用和语言结构的内在联系上。对于中国这样的社会，这些联系有着非凡的意义，可以说，任何实用性语言研究都需要认真、系统地考虑这些联系。

最近，中国又出台了很多对语言和语言研究产生直接和显著影响的重要政策和措施。例如，“一带一路”方略已经吸引了许多语言学家和语言政策制定者的极大关注。另一项政策，我特别感兴趣并且认为对语言研究有重要意义，是计划生育政策，具体地说是城市地区放宽一对夫妇只生一个孩子的政策。

始于20世纪50年代的独生子女政策是全面计划生育政策的一部分。这项政策在20世纪80年代初的三、四年里被严格执行。单就生育控制方面，这项政策似乎取得了明显的成功：世界银行的数据表明1990年至2000年间，中国的人口出生率从2.61（每育龄女性）降到1.45，低于大部分西方发达国家。据人口学家和中国问题观察家的估计，中国在1980年至2010年间少生了2至4亿人。

该政策带来了广泛而深远的影响。据联合国统计，这项政策导致中国家庭“4-2-1”结构的出现。在中国，有三分之一的人口超过60岁，只有约48%的人口处于劳动年龄，老龄化现象非常严重。据美国中央情报局出版的《世界概况》，中国人口性别比例为119：100（男/女）。这些给经济（福利）、国家安全（军队）和文化（汉族与少数民族；城市和农村） 带来严重后果。

我主要关注的是一对夫妇生育一个孩子的政策对语言造成的影响，包括对语言使用和语言结构产生的影响。这是一个尚未系统研究而非常值得关注的领域，目前既是对比展开实证研究的最佳时间。调查的重点首先是儿童语言习得和语言发展。很多语言学家认为人类先天就有学习语言的能力，语言输入和学习环境对儿童语言习得和发展至关重要。我们同样知道这样一个事实，即孩子从照顾者那里获得的关注度非常重要，从他们那里获得的语言输入模式也同样重要。我们也知道，同伴互动对儿童成长非常关键，此外，性别也起着一定的作用。曾经有研究者关注过中国传统的育儿习俗，如将孩子背在母亲的背上而不是将他们抱在前面，可能会减少孩子和作为主要看护者的母亲之间的“共同注意”，也因此降低了对语言发展可能相当重要的输入的质和量。现在，作为家里的独生子，一个典型的中国孩子在家庭里可能得到了过多的关注。他们有大量的刺激物，如玩具、书籍、电视、电子产品、游戏等陪伴左右。但这并不意味着他们会有更多的“共同注意”，尤其是和父母的“共同注意”相对较少。事实是，他们接收到的语言模式在很大程度上是“成人”的，甚至是“老人”的，因为有祖父母参与其中。目前儿童的语言发展缺乏“同胞模式”，即没有兄弟姐妹之间的互动，同胞互动与同伴互动有着根本性差异。这些现象与社会文化共同促使年幼的孩子出现明显的早熟现象。我们在电视上看到这样的例子，即除了一些智力竞赛（如数学、常识，我只对语言有兴趣），幼儿会因为他们能够背诵上百首唐诗而得到奖励。结果“小大人”常被看成儿童发展的成功典范，这在他们的日常语言行为中表现得尤为突出。“过度保护”“溺爱”或培养孩子的超前意识/目标，都会对语言产生严重的后果，对此我们需要进行系统的研究。

同胞互动的缺失带来的语言影响值得特别关注。虽然有足够的同龄人之间的交往，但这种交往属于“陌生人间的互动”，在性质上是非家族的、非亲属的。表面上看，他们缺少了非统称形式的亲属称谓（如姑姑、舅舅等统称形式的，如叔叔、阿姨等和这些不同）。我们还注意到，在幼儿园和校园环境，孩子们与非家庭成员的同龄人间交流的话题通常与家庭无关，几乎不讨论日常家庭生活。他们只能从大人那里获取家庭生活方面的语言输入。这必然会产生一定的长期影响。孩子们在幼儿园和学校那里得到的输入模式往往来自于阅读/文字材料、媒体和成人。我们有必要对此进行更深入的研究。

另一个独生子女政策对语言产生特殊影响的领域或议题是隔代教养。我对隔代教养的关注已有一段时间，主要是因为我的研究重点在海外中国人社区。我们从过去的语言研究中得知，祖父母的角色在语言传播、语言保持和语言转换上至关重要。祖父母往往在保持本族语言、少数民族语言、 家庭语言或继承语言方面起到关键作用。在海外的中国家庭中，祖父母经常是中文输入的主要来源，其中包括方言。我们曾对说客家话的家庭进行了研究，结果显示，随着祖父母的去世，其语言也随之消亡，家人最终经历了语言的转变。在当代独生子女的中国家庭里，由于父母平时忙于工作和赚钱，爷爷奶奶在养育子女方面发挥着关键作用。在中国，祖父母以及成年陌生人（保姆）在育儿方面的参与度比世界任何一个地方都要多，这在一些西方国家是闻所未闻的。大多数城市的中国家庭由保姆和爷爷奶奶一起照顾孩子。有趣的是，方言传承并没有成为中国城市家庭生活的常态。即使爷爷奶奶可能是方言使用者，并且保姆也可能是，但小孩讲的话通常仍是标准的或者至少是超地方的语言（超地方方言标准是方言学、历史语言学和社会语言学研究的一个重大课题）。当前，人们对汉语标准语转换普遍持积极态度。它可能不一定是普通话，而是一些地方标准语，但绝对不是地方方言。同时，人们对幼儿进行的所谓双语教育也产生了极大的兴趣，即让他们学习英语，而且相信越早接触英语越好。

不过，祖父母对儿童的影响，特别是对儿童语言发展的影响不仅在方言的保存上面。这还与老年人对年幼孩子的讲话方式、词汇的选择和语法结构有关。语言的变化主要通过代际交流发生。祖孙之间更多互动的结果可能是两个非常不同的发展方向——聚合或离散。当两代人相互学习并适应对方时会发生聚合，反之则离散。这是一个非常有趣的话题，值得继续探索，该话题对某种语言的结构变化也会造成实质性影响。

还有许多议题和问题值得探讨。例如，夫妻双方都是独生子女，他们之间的互动（“独二代”之间的交流）问题（与儿童语言发展无关，但关乎成人语言的使用），不受独生子女政策影响的少数民族语言之间的比较，城市和农村之间语言的比较等，都值得探讨。

随着独生子女生育政策的放宽，我们会看到一批二孩出现，有些二孩在第一个孩子出生10年后才出生。这样的年龄差距会对孩子的语言发展和语言行为产生什么样的影响，将成为未来研究的一个重要课题。

我上面所谈到的研究是一个正在发展的研究领域，即“人口语言学”的一部分。人口语言学一方面关注人口的结构和变化之间复杂的相互关系，另一方面聚焦于人口结构和语言发展、语言变化、语言保持、语言转变、语言规划之间的关系。这不同于“语言人口学”，它主要研究语言分布在统计学上的解释。

在过去的二十年中，人口语言学已成为人口学和语言学的国际性交叉学科。有些人甚至认为近期语言景观方面的研究与其有所关联或者是其延伸的内容。人口语言学研究往往侧重于家庭等私人领域内语言的时空分布和动态；语言景观研究，从字面意义上讲，更关注公共领域，即街头巷尾可见的语言。在这个意义上，语言景观研究的结果不应该被视作某地区人口语言构成的事实依据。

人口语言学研究的一个关注重点是某些语言而非“民族”语言的活力问题。在特定语境或视角下，这类语言常指少数民族语言或主流语言。因为数字分类与社会分类不一定重合，因此，对这些语言感兴趣的研究人员讨论数字但又不仅限于数字，除数字外还需考虑权力和意识形态问题。

在我进行的人口语言学研究中，我对人口的统计分布不太感兴趣，这虽然是人口学作为一个调查领域的重要组成部分，但不是唯一的部分。我更感兴趣的是人口素质和在特定社区或社会中的人口构成的本质，即人口动态性。这是人口学研究的新潮流，可以用于分析任何形式的现居住人口的特征，即在时间和空间上的变化。它涵盖了人口的规模、结构、分布，以及因出生、迁移、老龄化、死亡所引起的空间或时间上的变化。的确，它往往会涉及定量问题——人口统计学，但人口学家当今所谈论的大多数问题实际上都是与人口动态有关的社会变化。许多大学把人口统计学视作社会学的一部分，因为社会学关注民族、宗教、种族和社会经济阶级。有些人将形式/结构人口学（或人口统计学）与社会人口学或人口研究加以区别，形式/结构人口学主要关注人口演变过程的测量，社会人口学或人口研究则着眼于影响人口的经济、社会、文化和生物这些进程之间的关系。

社会语言学及应用语言学中，许多探讨人口和语言关系的问题，逐渐发展成为人口语言学。在我所感兴趣的研究领域中，双语制和多语制、语言保持和语言转变、语言流失和语言死亡是人口语言学的中心议题。这些问题显然与某些语言使用者的数量和人口变化，以及对语言使用、语言结构和语言命运的变化有关。还有许多学者和政策制定者对在语言教学中可能成为全球语言的中文传播感兴趣。因此，他们对人口语言学充满兴趣，并且特别关注人口动态——谁是学习者，他们分布在哪里，以及如何让汉语教学在某些国家得到可持续性发展。汉语使用者遍布世界各地，但主要是华裔。与阿拉伯语、英语、法语、西班牙语、俄语，甚至和许多其他语言相比，将中文作为其主要语言的非华裔人口数量十分有限。有鉴于此，我们是否应该将中文作为继承语言和作为国际性语言加以区别？

如前所述，我感兴趣的不是简单的数字，而是人口构成和人口动态的本质。例如，老龄化似乎是发达国家的一个主要社会现象，尽管婴儿死亡与少女怀孕仍是非洲和其他欠发达或发展中国家的共同现象。由这些现象导致的语言后果是什么？在不同国家和不同语言文化背景下，不同世代间隔的意味着什么？ 例如，与一个世代间隔为35年或更长的社区相比，一个世代间隔小于20年的社区，对语言的影响是什么？从语言隔代研究的结果我们了解到，不同年龄段语言使用者所使用的语言是不同的。那么，不同世代间隔是否导致语言结构变化呢？世代间隔的长短是否会影响语言变化的快慢呢？这些问题都还没有答案，但都是非常有趣且重要的问题。混合语言婚姻是另一个例子。哪种或哪些语言是经过世代传承的以及传承的方式如何，从长远来看，这是一个非常有意义的重要议题。

中国新的政策措施和由此带来的人口变化为我们研究其对语言的影响和进一步发展人口统计学提供了非常难得的机遇。我衷心希望，中国的研究者能够抓住这一机遇，开展高质量的实证研究，所有这些将具有深远的理论意义和实践意义。

（北京外国语大学 张天伟、李艳红译）

Language and Demography： Implications of Chinas Family Planning Policies and Demographic Changes for Language and Linguistics

University of London Li Wei

Since 2005， the Chinese authorities have been com?mis??????sioning a series of reports on language use in social life in China. This has been a major development in the study of language in China and has attracted a great deal of attention overseas. The English language editions of selected articles from these reports are now available from De Gruyter. These reports refocused our attention to the intrinsic connections between social policies and what is happening in society broadly， on the one hand， and language use and language structures， on the other. For a society like that of the Chinese， these connections are of crucial significance， and any sensible and useful study of language needs to take them into consideration seriously and systematically.

More recently， there have been a number of key policy initiatives that have direct and significant impacts on language and studies of language. For example， the One Belt One Road initiative has already attracted a great deal of attention amongst linguists and language policy makers. Another policy initiative， the one that I am particularly interested in and the one that I feel also has important implications for language and language studies， is that of family planning， specifically the recent relaxation of the one couple one child policy in urban areas.

The one couple one child policy was part of a broader Family Planning （or birth control） policy that started in the 1950s. The strictest form of the policy was only ever implemented in specific areas of China for 3-4 years in the early 1980s. Purely in terms of birth control， the policy appeared to be a clear success. According to the statistics from the World Bank， Chinas national birth rate fell from 2.61 per woman of childbearing age to 1.45 between 1990 and 2000， much lower than many developed countries. Demographers and China watchers estimated that China prevented between 200 million and 400 million births between the 1980s and 2010.

The impacts of the policy are much more wide-ranging and fundamental. It has led to a 4-2-1 structure of Chinese families， with a severe ageing population of 1/3 over 60 years of age， and only about 48% of the population at working age， according to UN statistics. The World Factbook， by CIA， has the sex ratio of the Chinese population at 1.119：1 （male to female）.These have huge consequences for the economy （welfare）， national security （army）， and culture （ethnic minority vs Han Chinese birth ratio in the 2010 census was 11%-17% against national average of 8%； urban vs rural） of the nation.

I am of course primarily interested in the linguistic consequences of the One Couple One Child Policy， what impact the policy has had on the language practices and indeed the structures of the language. This remains an explored area. And it is time that we start doing some serious empirical research. The key area for investigation has to be child language acquisition and development. Whilst most linguists accept that human beings are pre-wired for language acquisition， input and environment play crucial roles in how language is acquired and developed in infant and childhood. We know for a fact that the amount of attention a child receives from the primary caregiver is important， so is the linguistic model（s） （the kind of input） they receive. We also know that peer interaction is equally crucial， where gender also plays a role. Once a upon a time， researchers were concerned that traditional Chinese child rearing practices， such as carrying the child on the mothers back， as opposed to carrying them on the front， might reduce the amount of joint attention between the child and the mother as the primary caregiver and therefore reduces the amount of input and the quality of the input which would be important for language development. Now a typical Chinese child in a one couple one child family may have too much attention， as they are the only child in the family. There is an abundant amount of stimuli – toys， books， television， electronic devices， games， etc. It does not mean that there is more ‘joint attention， certainly not always with the parents. But the fact is that the linguistic modelling they receive is largely ‘adult， even ‘elderly because of the grandparents involvement. There is a lack of sibling model， and no sibling interaction – sibling interaction is qualitatively different from that of peer （same age group） interaction. This is coupled with a societal culture that encourages apparent early maturity in young children. We see on TV， for example， young children are praised to be able to recite hundreds of Tang poems， apart from other intelligence-oriented competition （maths， general knowledge – Im only interested in language here）. So， ‘Little Adult （小大人）is often pre?sented as a successful model of child development， and it is particularly vivid in their everyday language behaviour. Issues regarding ‘over-protectiveness or even spoiling the child （娇惯）； the parents pushiness in developing the child ahead of time （‘超前意识/目标）， all have serious linguistics consequences that need to be studied systematically.

The specific linguistic effects of lack of sibling interaction deserves particular attention. There is sufficient peer interaction， but they are all ‘stranger interaction， with non-family/non-relatives. We know superficially that there is lack of use of non-generic kinship terms， such as 姑姑，舅舅， etc. But we notice that topics children talk about with non-family peers， usually in nursery and school contexts， are not domestic ones， i.e. not about everyday family life. They only have input about the language of domestic life from adults. This must have some long-term consequences. The input models the children get in nurseries and schools tend to come from reading/print material， the media， and in adult forms. We certainly need to look into it more.

Another area or topic where the one couple one child policy has had a distinctive impact on language is grandparenting. I have been interested in grandparenting for some time， mainly because of my work in the overseas Chinese communities. We know from many generations of linguistic research that the role of grandparents is crucial in language transmission， language maintenance and language shift. Grandparents often hold the key role in keeping alive regional dialects or ethnic minority/home/heritage languages. In the overseas Chinese families， grandparents are often the main source of input of Chinese， including Chinese dialects， and we have studied Hakka-speaking families where the language died with the grandparents and the family subsequently experienced a language shift. In contemporary Chinese families with one child only， the grandparents also play a key role in child rearing， simply because the parents are usually very busy working and making money. It is a lot more common in China， than anywhere else in the world to involve grandparents， as well as adult strangers （阿姨‘nannies）， in child rearing – unheard of in some western societies. Most urban Chinese families have nannies as well as grandparents looking after the child. Intriguingly， however， dialect transmission does not appear tobe the norm in urban Chinese families. So， even though the grandparents may be dialect speakers， and the nannies might be too， the speech norm that is used with young children is typically standard， or at least supra-local （supralocal dialect levelling is a major topic of research in dialectology， historical linguistics and sociolinguistics）. There is an overwhelming positive attitude towards shifting to standard varieties of Chinese. It may not always be Putonghua. It could be some local standard， but definitely not regional dialects. In the meantime， there is also great interest in the so-called bilingual education to very young children， i.e. teaching them English， and the popular belief that the earlier the exposure to English the better.

But the influence of grandparents on children and especially on the childrens language development is a lot more than just dialect maintenance. It relates to the way older adults speak， including their lexical choice and grammatical structures， to young children. Language change takes place primarily through inter-generational interaction. The consequences of more interactions that involve grandparents and grandchildren may have two very different directions – convergence， or divergence. Convergence when the two generations learn from each other and accommodate to each other. Divergence when they dont. This is an extremely interesting topic to explore further， a topic that has real substantial consequences to the changes of the linguistic structures of the language.

There are many other topics and questions that could be investigated. For example， interactions between couples who are single child themselves （‘独二代之间的交流）. This is not about child language development， but about adult language use. Comparisons between the so-called ethnic minority languages in China who are not directly affected by the one couple one child policy with the Han Chinese， and between rural and urban areas， are also worth doing.

With the new relaxation of the One Couple One Child policy， we are predicting to see a new group of second children， born some 10 years after the first. How such age gaps will impact on the language development of the children and the language practices within the family will become an important topic for future research.

The research I have suggested above is part of a developing field， known as ‘demographical linguistics. Demographical Linguistics concerns the complex inter?relationships between demographic patterns and changes on the one hand， and language development， language change， language maintenance and language shift， and language planning， on the other. It is different from ‘linguistic demography – statistical account of the distribution of languages.

Over the last twenty years， Demographical Linguistics has become an international crossing for demography and linguistics. Some even see the more recent develop?ments in the so-called linguistic landscaping research as connected with it or an extension of it. Whereas demo?graphical linguistic studies tend to focus on the spatial and temporal distribution and vitality of languages in the private domains of the home， linguistic landscaping has as its focus the public domain in the most literal sense， i.e.， in terms of the visibility and display of languages on the streets. In this sense， the outcomes of linguistic landscaping research shouldnt be taken as the facts of the linguistic make-up of the population in a given place.

One of the key concerns of demographical linguistics is the vitality of languages other than “national” languages. Depending on particular contexts or perspectives， such languages are often referred to as minority languages or dominated languages. Numerical classifications do not necessarily coincide with social classifications. So researchers who are interested in these languages talk about numbers but are not solely concerned with numbers. There are issues of power and ideology that need to be considered.

In my studies of demography and linguistics， I am less interested in the statistical distribution of the population， which is an important part but not the only important part of demography as a field of inquiry. I am more interested in the quality of the population， and nature of the composition of the population in specific communities or societies – in other words， population dynamics. And that is at the heart of a new wave of demographic research， to analyze any kind of dynamic living population， i.e.， one that changes over time or space. It encompasses the study of the size， structure， and distribution of these populations， and spatial or temporal changes in them in response to birth， migration， ageing， and death. Yes， it often talks in quantitative terms – the demographics， but most of the questions demographers are asking nowadays are actually about social change in relation to population dynamics. Many universities treat demography as part of Sociology because of its interests in nationality， religion， ethnicity， and socio-economic class. Some people make the distinction between formal/structural demography （or demographics） which is primarily concerned with the measurement of population processes， and social demography or population studies which focus on the relationships between economic， social， cultural and biological processes influencing a population.

Many topics in socio- and applied linguistics deal with the relationship between demography and language， which give rises to Demographical Linguistics. In the areas that I am particularly interested in， which are bilingualism and multilingualism， language maintenance and language shift， language loss and language death are all central topics in this field. They are clearly related to the number of speakers of particular languages and how demographic changes affect language use， the structures of the languages and the fate of the languages. Many scholars and policy makers alike are interested in the spread of Chinese as a possible global language through language teaching. They should therefore be interested in demographical linguistics and pay specific attention to population dynamics： who the learners are， where they are and how sustainable Chinese language teaching is in specific countries. Chinese speakers are found all over the world， but mainly amongst ethnic Chinese people. The number of people who are not ethnic Chinese but who speak Chinese as their primary language is extremely limited， compared to Arabic， English， French， Spanish， even Russian， and many other world languages. Should there be a distinction between teaching Chinese as a Heritage language and Chinese as an international language？

As I said earlier， what I am interested in is not simply the numbers， but the nature of the composition of the population or population dynamics. For example， ageing seems to be a major social phenomenon in developed countries， though infant death and teenage pregnancy are still a common phenomena in Africa and other under-developed or developing countries. What are the linguistic consequences of these？ What are the implications of different generation intervals in different countries and cultures for language – for instance， what are the linguistic consequences of one community whose generation intervals are under 20 years， compared to， say， another community whose generation intervals are 35 or over？ We know from generations of linguistic research that different age groups of speakers use language differently. Will the language structures change as a result of the different generation intervals？ Will linguistic change be faster or slower if the intervals are shorter or longer？ These are all questions with unknown answers right now. They are extremely interesting and important questions. Mixed language marriage is another example. What language or languages are being transmitted through the generations and how， is an important topic with long-term implications.

The new policy initiatives and population changes in China provide us with a unique opportunity to study their impact on language and subsequently contribute to the further development of Demographic Linguistics. I sincerely hope that researchers in China will seize this opportunity and conduct high-quality empirical research that will have long term theoretical and practical significance.