該培養雙語兒童嗎？幾個實際的考量
Bilingualism has always been a hot topic in Taiwan, but even more so today with the government’s Bilingual 2030 policy and the growing emphasis on English education. As someone who uses Chinese and English professionally, researches language and education, and is raising a young child bilingually, I approach the subject from both scholarly and personal perspectives.
長期以來，雙語教育一直是臺灣社會關注的議題。近年來，隨著政府推動「2030雙語政策」，以及大眾對英語教育日益重視，相關討論也更加熱烈。除了在工作中同時使用國語與英語研究語言及教育之外，我也正在以雙語的方式養育兒子。因此，我希望從學術研究與個人經驗兩個角度來討論這個議題。
My son is two and a half years old. I speak to him predominantly in English, while my wife speaks to him mostly in Mandarin Chinese. Because our family moves between Taiwan and North America, we want him to participate meaningfully in both environments.
我兒子目前兩歲半。我主要以英語跟他互動，而我的妻子則多使用國語。由於我們時常往返於臺灣與北美之間，因此我們希望他能自在地融入這兩種語言環境。
Like many parents, I initially assumed that bilingualism might improve his cognitive flexibility. However, current research does not support the claim that childhood bilingualism will automatically result in broad cognitive advantage.
我跟許多家長一樣，最初也認為雙語經驗或許有助於提升孩子的認知彈性。不過，目前的研究並不支持「兒童學習雙語必然帶來廣泛認知優勢」的說法。
In fact, recent meta-analyses have found little evidence that bilingual children consistently outperform monolingual children in executive functions such as attention, inhibition, working memory, and task switching. The average differences were very small and varied substantially across studies. After researchers adjusted for publication bias, meaning that studies reporting positive effects may have been more likely to appear in academic journals, some of the apparent bilingual advantages were no longer statistically reliable (Gunnerud et al., 2020; Lowe et al., 2021).
事實上，近年的後設分析發現，尚不足以證明雙語兒童在注意力、抑制力、工作記憶及任務切換等執行功能上，能持續優於單語兒童。兩者的平均差異非常小，而且不同研究所得出的結果也有很大差異。研究人員在校正出版偏誤後，也就是考量「發現正面效果的研究可能更容易獲得發表」這項因素後，原本看似存在的部分雙語優勢便不再具有統計上的可靠性（Gunnerud et al., 2020; Lowe et al., 2021）。
That does not mean bilingual experience has no cognitive significance. And when I refer to bilingualism here, I am not limiting the discussion to Mandarin and English. Under certain conditions, bilingual experience may be related to particular forms of language awareness or cognitive processing. But parents should not expect bilingual education to automatically make their children smarter, more attentive, or more academically successful.
這並不代表雙語經驗完全沒有認知上的意義，而且這裡所說的雙語也不只限於中文與英文。在特定條件下，雙語經驗可能與某些語言意識（language awareness）或認知處理方式有關。不過，家長本來就不應期待雙語教育能自動讓孩子變得更聰明、更專注，或有更好的學業表現。
For my family, the strongest reasons for bilingualism are practical and relational. Chinese and English allow our son to communicate with different family members, participate in life in Taiwan and North America, develop literacy in two languages, and preserve future educational and professional options. These benefits are meaningful, but they are neither automatic nor cost-free.
就我們家而言，培養雙語能力最重要的原因，其實更貼近日常生活與人際關係。國語與英語能讓兒子直接與不同的家庭成員交流，融入臺灣與北美的生活，並發展兩種語言的讀寫能力。這些好處確實具有意義，但不會自動出現，也不是毫無代價。
So, is bilingualism worth the effort?
那麼，培養雙語能力值得嗎？
It helps to look at both the advantages and the disadvantages before deciding what makes sense for your own family. In this article, I will start with some of the potential benefits.
我們可以先分別檢視雙語發展的優點與缺點，再思考什麼樣的選擇最適合自己的家庭。這篇文章先從幾項可能的優點談起。
Family Communication 家庭溝通
As I stated earlier, the clearest benefit is the ability to communicate directly with family members who use different languages. For Taiwanese families living abroad, maintaining Mandarin, Taiwanese Hokkien, Hakka, or another family language can help children communicate more independently and precisely with parents, grandparents, and relatives.
如上所述，最直接的好處之一，就是孩子能夠直接與使用不同語言的家人交流。對居住在海外的臺灣家庭而言，維持國語、臺語、客語或其他家庭語言，有助於孩子更獨立且精確地與父母、祖父母或其他親屬交流。
Sharing a language does not guarantee emotional closeness, but it expands what family members can talk about without relying on translation. A child may understand basic everyday instructions in a language but still be unable to discuss emotions, values, family history, conflict, or important personal decisions in that same language.
擁有共同語言並不能保證家人之間一定親密，但能擴大彼此可以直接討論的內容，而不必依賴翻譯。孩子即使聽得懂日常生活中的基本指令，也未必能用同一種語言討論情緒、價值觀、家庭歷史、衝突或重要的個人決定。
One study of Chinese American children between the ages of seven and eleven found that the amount of Chinese used by adults in the home predicted stronger receptive and expressive Chinese vocabulary about two years later. Participation in Chinese-language extracurricular activities also predicted stronger Chinese vocabulary and word-reading skills.
一項針對七至十一歲華裔美籍兒童的研究發現，家中成人使用中文與孩子互動的程度，可以預測孩子約兩年後較強的理解性（receptive）與表達性中文詞彙能力。孩子參與中文課外活動的程度，也能預測其中文詞彙與認字表現。
Interestingly, simply having positive attitudes toward Chinese or allowing children to passively consume Chinese-language media did not predict better Chinese-language outcomes (Chen et al., 2021).
值得注意的是，家長只是對中文抱持正面態度，或只是讓孩子被動接觸中文媒體，本身並不能預測較好的中文學習成果（Chen et al., 2021）。
These findings should not be generalized directly to every Taiwanese family or every age group. Still, they point to an important principle: children need regular opportunities to hear and actually use a language. Ideally, they also need opportunities to express meaningful and increasingly complex ideas through it.
這些研究結果不能直接套用到所有臺灣家庭或所有年齡層，但仍說明了一項重要原則：孩子需要規律地接觸並實際使用該語言，而且最好能有機會透過這種語言進行深入、有意義且逐漸複雜的表達。
Participating in Two Educational and Cultural Environments 參與兩種教育與文化環境
Another benefit is that children who can actually use both Mandarin and English may find it easier to participate in education, relationships, and everyday life in both Taiwan and English-speaking countries. Bilingual or multilingual ability may also provide them with more options for future study, travel, immigration, and professional communication.
另一項好處是，能夠實際運用中文與英文的孩子，可能更容易參與臺灣與英語系國家裡的教育、人際關係與日常生活。雙語乃至多語能力，也可能為未來的求學、旅行、移民與專業溝通保留更多選項。
In Taiwan, English also has a very practical role in education and employment. It is part of the school curriculum and educational assessments, and some companies and specific jobs do require or consider scores from English proficiency tests such as TOEIC or IELTS. English ability is therefore useful, even when a child does not go abroad. It may also directly affect educational opportunities and some employment opportunities within Taiwan.
在臺灣，英文也確實具有實際的教育與求職功能。英文是學校教育與升學評量的一部分，而部分企業與特定職缺也會要求或參考 TOEIC、IELTS 等英語檢定成績。因此，英語能力不只是未來出國時才有用，也可能直接影響學生在臺灣的升學選擇與部分求職機會。
At the same time, these are possibilities rather than guarantees. Being good at English does not mean that a child will automatically get into a particular university, find a better job, or earn a higher salary. Pan and Roever (2016) interviewed employers representing seventeen Taiwanese companies. They found that English certificates were generally not the primary factor employers considered when hiring. Some employers saw test scores as useful indicators of language ability, but they also pointed out that intensive test preparation can produce strong scores without necessarily reflecting a person’s ability to communicate effectively in real-world situations.
不過，這些都只是可能性，而非保證。英語能力不代表孩子一定能進入某一所大學、找到更好的工作，或獲得更高收入。Pan 與 Roever（2016）訪談了十七家臺灣企業的雇主。研究發現，英語證照通常不是企業聘用員工時的首要考量。有些雇主認為證照可作為語言能力的參考，但也指出應試訓練可能讓求職者取得不錯的成績，卻未必能反映真實情境中的溝通能力。
So while English proficiency and test scores may have real value for education and employment in Taiwan, the long-term goal should not simply be a high score or another certificate. It should be the ability to actually learn, work, and communicate effectively when English is needed.
因此，英文能力與檢定成績在臺灣確實可能具有升學與求職上的現實價值，但長期而言，真正重要的仍不是單純取得高分或證照，而是能否在需要使用英文時，有效地學習、工作與溝通。
Does AI Change the Value of Learning to Read and Write in Two Languages?
人工智慧會改變接觸兩種讀寫系統的優勢嗎？
Biliteracy is one of the main reasons I support bilingual education. But as artificial intelligence develops rapidly, we need to be more precise about what that advantage actually means.
雙語讀寫能力，是我支持雙語教育的重要理由之一。不過，隨著人工智慧快速發展，我們需要更精確地界定這項優勢究竟代表什麼。
Yes, AI can now translate and summarize texts quickly, making books, articles, and educational materials more accessible to people who do not know the original language. As these tools improve, children may not need advanced proficiency in two languages merely to understand the general content of a text.
人工智慧現在已經能快速翻譯與摘要文本，讓不熟悉原文的人也能接觸書籍、文章與教育資源。隨著這些工具持續進步，即使孩子的雙語能力沒有達到高度流利的程度，也可能藉由 AI 大致理解文本內容。
However, translated access is not the same as direct literacy. AI systems can still misinterpret grammar, idioms, tone, cultural references, and meanings that depend on context. Studies of Chinese-English machine translation continue to find differences in performance across grammatical features, subject areas, and culturally specific content (Song et al., 2024; Yuan et al., 2026).
然而，透過翻譯取得資訊，並不等於具備閱讀原文的能力。人工智慧仍可能誤解文法、慣用語、語氣、文化指涉，以及必須結合前後文才能理解的文意。針對 AI 中英翻譯的研究顯示，系統在處理不同文法特徵、專業領域與特定文化內容時，表現仍有所差異（Song et al., 2024; Yuan et al., 2026）。
Children who can independently read in both languages are better able to understand an author’s wording, style, humor, ambiguity, and cultural assumptions directly. They are also better positioned to notice when an AI translation has simplified, distorted, or omitted part of the original meaning.
能夠獨立閱讀兩種語言的孩子，較能直接理解作者的措辭、風格、幽默、模糊語意與文化預設，也更有能力判斷 AI 的翻譯是否簡化、扭曲或遺漏了原文的某些部分。
AI has undeniably “weakened” one traditional advantage of biliteracy. Access to information is no longer limited only to people who can read the original language, but AI does not replace the ability to understand, interpret, evaluate, and communicate directly through two written languages (Oshchepkova et al., 2023).
人工智慧確實削弱了雙語讀寫的一項傳統優勢，也就是過去往往只有懂原文的人才能直接取得某些資訊。然而，它無法取代人們直接以兩種書面語言理解、詮釋、判斷與溝通的能力（Oshchepkova et al., 2023）。
Biliteracy, then, is not simply about knowing two writing systems. It gives children the ability to use those systems independently, make their own judgments, and avoid relying entirely on translation tools or AI. In fact, I would argue that biliteracy may be even more valuable in the age of AI translation, because as more people rely on AI to interpret texts for them, fewer may have the ability to judge whether a translation fully captures the original wording, tone, cultural references, ambiguity, and implied meaning. Being able to read both languages directly gives a person an independent basis for evaluating what the AI got right, what it simplified, and what it may have missed.
因此，雙語讀寫能力的價值並不只是「看得懂兩套文字系統」。更重要的是，它讓孩子能夠獨立使用這兩套語言系統並形成自己的判斷，而不必完全依賴翻譯工具或人工智慧。事實上，我認為在 AI 翻譯愈來愈普及的時代，雙語讀寫能力反而可能變得更加重要。因為當愈來愈多人依賴 AI 代為理解文本時，真正有能力判斷翻譯是否完整保留原文措辭、語氣、文化指涉、模糊語意與言外之意的人，反而可能愈來愈少。能直接閱讀兩種語言，就能有自己的判斷基礎，知道 AI 哪裡翻對了、哪裡過度簡化，又有哪些意思可能被遺漏。
Bilingualism Does Not Inherently Confuse Children
雙語本身不會讓孩子混淆
Strictly speaking, this is not a benefit of bilingualism. It is a clarification of a common misconception.
嚴格來說，這不算是雙語的優點，而是對一項常見迷思的澄清。
When children switch between languages within the same sentence or conversation, that does not mean they are confused.
孩子在同一句話或同一段對話中交替使用不同語言，並不代表他們混淆了這些語言。
Research reviews have concluded that exposure to two languages does not confuse children or interfere with normal language development. The two languages also do not need to be kept completely separate in every situation (Hoff & Core, 2015).
研究回顧指出，接觸兩種語言不會讓孩子產生語言混淆，也不會阻礙正常的語言發展。兩種語言也不需要在所有情況下完全分開使用（Hoff & Core, 2015）。
However, because bilingual children divide their language experience between two languages, their development in each individual language may appear slower when compared separately with monolingual children who receive almost all of their input in just one language.
不過，由於雙語兒童的語言經驗分布在兩種語言之間，如果只分別評估其中一種語言，他們在該語言上的發展可能看起來慢於幾乎所有語言輸入都集中在單一語言上的單語兒童。
This does not mean that bilingualism is harming their development. It is a normal consequence of distributing language experience across two languages.
這並不代表雙語阻礙了孩子的語言發展，而是語言經驗分布在兩種語言中的正常結果。
Bilingual children may also use their languages differently depending on the person they are speaking with, the setting, and the topic. Looking at only one language in one context therefore may not reflect their overall language ability.
雙語兒童使用哪一種語言，也可能根據談話對象、情境與主題而有所不同。因此，如果只根據其中一種語言或某一特定情境來評估，就可能無法完整反映孩子整體的語言能力。
So what are the takeaways?
所以，我們今天的文章到底有什麼重點呢？
- Bilingualism does not reliably make children smarter, more attentive, or more academically successful.
雙語未必能讓孩子變得更聰明、更專注，也未必能提升學業表現。 - The clearest benefits are practical: communicating with family, participating in different societies, and preserving future educational and career options.
雙語最明確的優點在於實際用途，包括與家人交流、融入不同社會，以及保留未來的教育與職涯選擇。 - Positive attitudes toward a language, or passive exposure through media alone, are not enough to develop strong language ability. Children need long-term, regular, and meaningful opportunities to hear and use each language.
只對某種語言持正面態度，或只是讓孩子被動接觸媒體，並不足以發展穩固的語言能力。孩子需要長期、規律且有意義地聆聽並實際使用每一種語言。 - AI can translate information, but biliteracy still provides direct understanding, cultural judgment, and the ability to critically evaluate AI-generated translations.
人工智慧可以翻譯資訊，但雙語讀寫能力仍能提供直接理解、文化判斷，以及批判性評估 AI 翻譯的能力。 - Mixing languages does not mean a child is confused. Bilingual development does not inherently interfere with language development.
混合使用不同語言並不代表孩子產生混淆。雙語發展本身不會阻礙正常的語言發展。 - Bilingual children may be stronger in one language than the other, and their abilities should be considered across different settings and uses rather than judged through only one language.
雙語兒童可能較擅長其中一種語言，因此應綜合考量他們在不同人物、情境與用途中的語言表現，而不是只以其中一種語言來判斷整體能力。
In the next article, I will look at some of the potential disadvantages of bilingual development and the practical issues Taiwanese parents and children may need to consider. I will also share why I choose to speak English with my son.
下一篇文章，我會繼續討論雙語發展可能面臨的缺點，以及臺灣家長與孩子在現實中需要考量的問題。我也會分享為什麼我選擇用英文跟兒子互動。
參考文獻
Chen, S. H., Zhou, Q., & Uchikoshi, Y. (2021). Heritage language socialization in Chinese American immigrant families: Prospective links to children’s heritage language proficiency. International Journal of Bilingual Education and Bilingualism, 24(8), 1193–1209. https://doi.org/10.1080/13670050.2018.1547680
Gunnerud, H. L., ten Braak, D., Reikerås, E. K. L., Donolato, E., & Melby-Lervåg, M. (2020). Is bilingualism related to a cognitive advantage in children? A systematic review and meta-analysis. Psychological Bulletin, 146(12), 1059–1083. https://doi.org/10.1037/bul0000301
Hoff, E., & Core, C. (2015). What clinicians need to know about bilingual development. Seminars in Speech and Language, 36(2), 89–99. https://doi.org/10.1055/s-0035-1549104
Lowe, C. J., Cho, I., Goldsmith, S. F., & Morton, J. B. (2021). The bilingual advantage in children’s executive functioning is not related to language status: A meta-analytic review. Psychological Science, 32(7), 1115–1146. https://doi.org/10.1177/0956797621993108
Oshchepkova, E. S., Kartushina, N. A., & Razmakhnina, K. O. (2023). Bilingualism and development of literacy in children: A systematic review. Psychology in Russia: State of the Art, 16(1), 3–25. https://doi.org/10.11621/pir.2023.0101
Pan, Y.-C., & Roever, C. (2016). Consequences of test use: A case study of employers’ voice on the social impact of English certification exit requirements in Taiwan. Language Testing in Asia, 6, Article 6. https://doi.org/10.1186/s40468-016-0029-5
Song, H., Li, Y., Wu, Y., Liu, Y., Lin, J., & Xu, H. (2024). How grammatical features impact machine translation: A new test suite for Chinese-English MT evaluation. In B. Haddow, T. Kocmi, P. Koehn, & C. Monz (Eds.), Proceedings of the Ninth Conference on Machine Translation (pp. 1200–1221). Association for Computational Linguistics. https://doi.org/10.18653/v1/2024.wmt-1.117
Yuan, Z., Ye, Y., Feng, X., Li, B., Hong, Q., Lu, Y., Tu, D., & Qin, B. (2026). Culture-aware machine translation in large language models: Benchmarking and investigation. In M. Liakata, V. P. Moreira, J. Zhang, & D. Jurgens (Eds.), Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics: Volume 1, Long Papers (pp. 29636–29661). Association for Computational Linguistics. https://doi.org/10.18653/v1/2026.acl-long.1367