Mises Daily Articles
English for All, Freedom for None
"English should be the official language of this country" has become something of a mantra for the Republican Party, uttered to raucous cheers in nearly every Republican debate or speech. Moreover, today several organizations exist to lobby for English as the official language, including U.S. English, ProEnglish, and English First. Though the aims of these organizations and politicians vary, they all agree on one central objective: to make English the official language of the United States.
Advocates of official English, however, have given insufficient thought to what they mean by the terms they use, the unseen effects such policies entail, and whether the means they choose are in fact fit to attain the ends aimed at. No good praxeological analysis can do without addressing all three of these concerns, each best summarized by some of the great thinkers in liberal philosophy and economics: It was Voltaire who said "If you wish to converse with me, define your terms"; and it was Claude Frédéric Bastiat who first articulated the core principle of economic analysis in his famous essay "That Which Is Seen, and That Which Is Not Seen," later summarized by Henry Hazlitt in his book Economics in One Lesson:
The art of economics consists in looking not merely at the immediate but at the longer effects of any act or policy; it consists in tracing the consequences of that policy not merely for one group but for all groups.1
Finally, it was Ludwig von Mises who established the most important aspect of any praxeological analysis in his magnum opus, Human Action:
In this sense we speak of the subjectivism of the general science of human action. It takes the ultimate ends chosen by acting man as data, it is entirely neutral with regard to them, and it refrains from passing any value judgments. The only standard which it applies is whether or not the means chosen are fit for the attainment of the ends aimed at.2
It is important to recall these insights when treating social issues like language, because so much of the previous commentary on official English strays far from any sort of objectivity. Because language is so intimately tied up into our identities and how we perceive ourselves, even Austro-libertarians can forget Mises's methodological individualism when language is involved. So in the spirit of these great thinkers, let's take a praxeological look at what "English as the official language" really means.
Defining "the Problem"
Official English movements have been cropping up in the United States for longer than there's been a United States. The colonial statist Benjamin Franklin penned in 1753,
They come in droves.… Few of their children in the country learn English.… Advertisements intended for to be general are now printed in Dutch and English, the signs in Our Streets have inscriptions in both languages, and in some places only German, they began of late all their bonds and other legal writings in their own language, which (tho' I think it ought not to be) are allowed good in Our Courts.3
Later, in 1868, the Indian Peace Commission (whose policy suggestions were anything but peaceful) recommended that "schools should be established which children should be required to attend; their barbarous dialects should be blotted out and the English language substituted."4 And in 1907 President Roosevelt famously wrote, "We have room for but one language in this country, and that is the English language, for we intend to see that the crucible turns our people out as Americans, of American nationality, and not as dwellers in a polyglot boarding house."5
Most recently, in December of 2011, Republicans in both the House and Senate — including cosponsor Ron Paul — introduced HR 99 to declare English as the official language of the United States. Rep. Steve King (R-Iowa), who sponsored the bill, stated on the topic that "A nation divided by language cannot pull together as effectively as a people."
How then do we define the problem that official English legislation is supposed to solve? What is the crisis that has had so many people worried throughout history? It is necessary to answer this question if we are to assess whether the proposed solution solves this problem effectively. Ostensibly, the most pressing concern is that citizens aren't learning English, and that this deprives them of certain opportunities. In the last Republican debate in Florida, for example, Mitt Romney remarked, "We don't want people to be limited in achieving the American dream because they don't speak English."
But is it really true that immigrants and their children aren't learning English? It turns out this is just blatantly false. A study conducted in 1998 showed that 93.6 percent of all children of immigrants spoke English well or very well.6 As I detailed in my Mises Daily "Why Languages Die," this is the standard process of language shift. By the third generation, all trace of the heritage language is gone. If this were not the case, it would be difficult to explain why Dutch, German, Italian, or other major immigrant languages in the States haven't spawned lasting communities of monolingual speakers. Even places like Chinatown in New York City only retain a persistent heritage language presence because they are continually resupplied by new immigrants. But the children of those immigrants continue to learn English, and their children continue to never learn the heritage language in the first place, making them English monolinguals.
In his book You Are What You Speak: Grammar Grouches, Language Laws, and the Politics of Identity, which is filled with excellent praxeological insights on language politics, Robert Lane Greene explains the situation well:
It is, to put it simply, nearly impossible to raise a child in the United States without the child learning English; it would require isolation from the outside world bordering on child abuse. Children born in America, and even those arriving at a young age, inevitably pick up English.
Of course, members of this bridge generation will typically speak the old country's language, too, learning it in the kitchen from their parents. But it is the bridge generation's own behavior as parents that eventually, and inevitably, dooms the "heritage" language. Parents are not stupid, nor do they typically perform cruel tricks on their children. Parents of children in the United States know that wherever they live, however many bilingual signs surround them, however many phone menus offer "Oprima dos para español," and however many fellow speakers of their language surround them, their children will need English to survive and thrive in America.7
So much for immigrants not learning English. What then of other "problems" with multilingualism? One perspective that is often implicit in the arguments of official English proponents is that English will become a threatened language in America. But this is a strange insecurity to have. Do conservatives really fear that liberals will find a way to ban English? Unless such a thing were to happen, it seems likely that English will continue to be the language of primary interaction in the States, even if only a minority of people speak it as their native language. In many places of the world English has become a lingua franca, despite nobody speaking it as a first language. Even in the Republicans' worst-case scenario, the United States would hardly be any different. And should Spanish ever become the primary language of significant swaths of territory in the States, one only has to look as far as developed countries like Switzerland — which has no less than four official languages — to see that such multilingualism is largely unproblematic.
The most common charge leveled against multilingualism is that it is "inefficient." Why waste millions of dollars translating ballots, legislation, censuses, and other government documents into multiple languages when we can just publish it all in one? Is this not more efficient? However, this seemingly common-sense argument faces a serious praxeological objection, which is that "efficiency" is a purely subjective notion.8 What does it mean for something to be efficient? Efficient for whom? For the millions of people who do not speak English well, having a ballot translated into Spanish — which takes perhaps half an hour of a translator's time — is a far more efficient option than spending hundreds or thousands of dollars and hours on learning English. Part of why efficiency is subjective is because the notion of efficiency depends on the notion of costs, and cost is a psychic phenomenon, not a monetary one. So when the organization English First supports legislation to "lower healthcare costs by using English only and removing the translator mandate," they are interested only in costs to healthcare providers, and ignore costs to Spanish-speaking individuals — a classic broken-window fallacy. These advocates have failed to take heed of Bastiat and Hazlitt's important admonitions.
In the same vein of "efficiency," official-English advocates are concerned about an unnecessary reduplication of effort, but reduplication of effort is one of the cornerstones of a free market — we call it "competition." To claim that translating documents into Spanish is a wasteful reduplication of effort is no different than claiming that having multiple producers of shoes is a wasteful effort. It misses the crucial point that the only reason someone would reduplicate effort in a market is because they expect the benefits to exceed the costs. In the case of language, there are many people — namely, those who speak Spanish as a first language — for whom the additional cost in tax dollars is minor compared to the relative benefit. Of course, this argument is complicated by the fact that these language services are likely on net funded by somebody else's money. This is the insoluble problem of every government-funded service. What this problem reveals, however, is that the argument from "efficiency" is not really about efficiency at all but rather about preferences, namely, the preference as to how other people's money should be used.9
Another ideology undergirding conservative apprehensions about language is the misguided belief that languages and cultures are largely equivalent. Therefore, the rise of Spanish constitutes — so the logic goes — a decline in American culture. Yet Franz Boas, the founder of American anthropology, debunked this ideology over a century ago, and it's easy to see how. Imagine if you will that everybody in America started speaking Spanish overnight. Would that make us all less American? Would our values change, our cultural heritage become undone? Of course not. We'd simply be an America that speaks Spanish. Yet Spanish is vilified because it serves as a symbol of a perceived encroachment on the core of American values. Anecdotally, I find this very strange, since the majority of Hispanic immigrants I know work very hard so that their children can live better than they did; this drive to advance oneself through one's own efforts is a central tenet of American values. More importantly, even if it were true that American culture were on the decline, and Latin American culture on the uptick, official-English or even English-only legislation would hardly prevent this. A Latin American culture that speaks English is still in most respects a Latin American culture. Language is then, at least in this regard, entirely the wrong target, a symptom rather than a cause, and thus cannot be seen as part of "the problem" as seen by most conservatives.
A final problem pointed out by English-language advocates is the "lack of national unity" caused by multilingualism. But we have already noted that there are prominent and successful multilingual countries that do not seem to be particularly lacking in national unity. Furthermore, to deviate from our value-free assessment for a moment, it should be clear to any libertarian that "national unity" implies increased state centralization in an attempt to overcome economic-calculation problems, and thus should be opposed in principle. But the objective point is that these activists have fallen prey to a pernicious conceptual fallacy, what Benedict Anderson called "imagined communities" in his book by the same name. Simply put, this is the notion that a country is a single, culturally homogenous unit, and that its citizens exhibit certain common traits. However, this is nothing more than a classic fallacy of composition, and it runs afoul of Mises's various critiques of collectivism, most significantly his principle of methodological individualism, which denies that entities like "nations" and "society" exist other than as collections of individuals, who each act individually and toward distinct ends.10
This mistaken ideology was expressed by Newt Gingrich in a presidential debate when he asked,
How do you bond the country together? How do we create a unified country? To unify ourselves in the future, it is essential to have a central language that we expect people to learn and be fluent in.
HR 997 is in fact even titled the "English Language Unity Act." The ideology is also expressed in the text of HR 997 itself, which says, "Throughout the history of the United States, the common thread binding individuals of differing backgrounds has been the English language." In addition to the critiques already noted above, there are two other scenarios that illustrate the absurdity of such statements: First, there are billions of speakers of English all over the world, most of whom feel no fraternal bond towards the United States (in fact in many cases quite the opposite). While language is often an important mark of cultural unity, this is hardly always the case. Second, it would seem rather ridiculous to imagine that college students at Berkeley, Navajos in Arizona, and evangelicals in South Carolina share some mythical communal bond by virtue of living in an arbitrary set of forcibly imposed political borders called America. Anderson was quite correct to call such a community "imaginary."
An extreme example of how tightly bound language and nationalistic concepts of culture are perceived to be comes from this rant published at DailyPaul.com, expressing righteous indignation at 33 senators who voted against making English the official language of the United States: "Your vote reflects betrayal, political surrender, violates your pledge of allegiance, dishonors historical principle, rejects patriotism, borders on traitorous action."
The false perception of an imagined community is also clearly embedded in common claims like "The language of America is English, and so people should learn English to be a member of this country." This statement is not just false, but self-contradictory. If English were really the language of America, then it would hardly be necessary for people to learn it. A correct statement is to say that English is the first language of the majority of Americans but that there exist plenty of people for whom Spanish is the primary language of everyday use, or for whom English is a second language. For these individuals, living in communities where they can buy their groceries in Spanish, send their children to school in Spanish, and even interact with their local government in Spanish, English is far from being a necessity. It may be true that knowing English would be beneficial to these individuals, but this fact does not justify the statement that they ought to learn it, or that they even need it.
The official-language movement, then, for all the reasons given above, is merely another example of activists and the government conjuring up nonexistent problems just so that they can solve them. Let us now see what ends English-language proponents have selected to fix this non-problem.
Defining "Official Language"
What do advocates mean when they suggest "making English the official language"? There seem to be two ends of a spectrum — those who focus primarily on seeing English adopted as the language of the federal government, thus eliminating "waste," and those who would like nothing more than to eradicate every language other than English. The English First organization appears to be closer to the latter end of this spectrum, noting proudly that it lobbied against bilingual ballots and bilingual education. Likewise ProEnglish tells us that "Official English would also reinforce America's historic message to new immigrants — that we expect them to learn English as the first step in their assimilation."
Others like U.S. English go to some lengths to clarify that they do not advocate "English only," merely English in the government. Nominally, they attempt to adopt a private-property stance. U.S. English states, for example, that "Making English the official language of the United States refers solely to the language of the government, not of the people, private business, classrooms, etc." They further clarify that "U.S. English, Inc. has never and will never advocate for any piece of legislation that bans the use of languages other than English within the United States" — unless, of course, that use is to interact with a government official. Such advocates do not see this contradiction in their own terms.
The actual legislation on this issue, HR 997, falls closer to the first approach, but has elements of the latter as well. The primary focus of the legislation is to declare that "The official language of the United States is English." On its face, this seems innocuous enough. English is, after all, the de facto language of the US government already. Such a gesture would seem largely symbolic, but this would be a dangerously naïve misconception. As Jeffrey Tucker astutely quips, "all laws have teeth." Immediately following the above proclamation comes the following statement:
Representatives of the Federal Government shall have an affirmative obligation to preserve and enhance the role of English as the official language of the Federal Government. Such obligation shall include encouraging greater opportunities for individuals to learn the English language.
The document also notes that "the term 'official' refers to any function that (i) binds the Government, (ii) is required by law, or (iii) is otherwise subject to scrutiny by either the press or the public."
The immediate conclusion here is that official-English advocates themselves have not adequately or precisely defined the policies they propose, with the result that official-English legislation is so vague that its scope could be interpreted without limit, much like the commerce clause of the Constitution. Jeffrey Tucker, in his wonderful book It's a Jetsons World: Private Miracles and Public Crimes, brilliantly captures the function of such vagueness in legislation:
When dealing with the state, we need certain rules, not grey areas — the less discretion the better. But the state doesn't like certainty. There is an obvious advantage to vagueness for the government. It keeps everyone living in a state of fear. The arbitrariness of it all makes us nervous and constantly aware of who or what is in charge.
But to what end? It's not like [the free market], where rule ambiguity is concocted with the final goal of serving us. When the state creates legal ambiguity — and it does so with deliberation — it is for the purpose of allowing them to trap us, tax us, coerce us, keep us on edge and living in fear.11
To add to this confusion, HR 997 also lists numerous "common-sense" exceptions to the legislation, including
- teaching of languages
- special considerations for individuals with disabilities
- actions, documents, or policies necessary for national security, international relations, trade, tourism, or commerce
- actions or documents that protect public health and safety
- actions or documents that facilitate the census
- actions that protect the rights of victims of crimes or criminal defendants
- using terms of art or phrases from languages other than English
- members of the government from communicating un-officially in other languages
- the preservation or use of Native American languages
- any person learning a language
As Per Bylund points out, all these exceptions are an attempt by the government to approximate "market-like" solutions to otherwise simple problems. Rather than simply letting the preferences of individuals determine how much of a certain good will be present on the market — in this case, who speaks a certain language, how often it's spoken, and to whom — a bewildering array of rules and exceptions is put in place instead. I believe it's also worth pointing out that in practice, once all the various exceptions are taken into consideration, really only two activities are vigorously targeted by this law — voting in a foreign language, and translating "nonessential" government documents. As there is no mention of bilingual education in this legislation, it can also be assumed that HR 997 would also be interpreted to prohibit that activity at some point in the future as well.
Thus the answer to the question of "How do official-English supporters propose to solve the non-problem of multilingualism?" is that they advocate dangerously vague prescriptions that create unnecessarily complex solutions and burdensome regulations, with no discernible direct actions to be taken. Their solution, like their problem, is a non-solution — a mere prohibition on the actions of individuals. This is precisely analogous to minimum-wage laws, which don't in fact create a minimum wage at all; rather, they simply prohibit any activity that falls under some arbitrary minimum number. In the same way, official-English laws merely prohibit a certain kind of arbitrarily chosen activity.
That Which Is Not Seen
It should hardly be necessary at this point to list the unseen effects of official-English policies, as so many of those effects were brought to light even in our initial analysis. Yet there is still much more that can and should be said. I will note a few final points here.
By far one of the biggest mistakes that official-English advocates make in their economic reasoning is that they neglect the influence of the past on action. This insight is of course due to Ludwig von Mises, whom I quote here in somewhat modified form:
We may picture to ourselves the image of how things would be if, equipped with our present knowledge of natural resources, geography, technology, and schooling, we had arranged all the languages of the world accordingly.… The writings of the socialists are full of such utopian fancies. Whether they call themselves Marxian or non-Marxian socialists, technocrats, or simply planners, they are all eager to show us how foolishly things are arranged in reality and how happily men could live if they were to invest the reformers with dictatorial powers.
The fundamental error involved in this rationalistic romanticism is the misconception of the character of capital goods available and their scarcity. The languages spoken today were passed down by our ancestors and ourselves. The plans which guided who learned what language and who taught what language to whom were an outgrowth of the then prevailing ideas concerning ends and the usefulness of various languages. If we consider aiming at different ends and choosing different languages to speak, we are faced with an alternative. We must either leave unused a great part of the linguistic knowledge acquired by individuals and start afresh with new education, or we must adjust our educational processes as far as possible to the languages which are actually spoken. The choice rests, as it always does in the market economy, with the consumers.
The fact that English is not instantly adopted by every speaker is not more conspicuous than the fact that not everybody throws away his old car or his old clothes as soon as a better car is on the market or new patterns become fashionable. In all such things people are motivated by the scarcity of goods available.12
In the particular case of language, decisions on which languages to learn and to teach are motivated by the scarcity of educational resources available, and their perceived potential future value. In their utopian fantasizing, English advocates hear the phrase "English as the official language" and envision a world where everybody magically speaks English, businesses and government agencies are required to conduct transactions in English, and all this happens at no cost whatsoever, much as left-liberals imagine a world where everybody has "free" healthcare without wondering who pays for it. Not only does education for ESL students require special training and funding, but poorly educated and immigrant populations are precisely those that are least capable of spending the additional time and money to acquire English, thus disenfranchising the very individuals that official-English legislation is supposedly meant to help. Therefore, in their own terms, the results of policies advocated by official-English proponents are a failure, and inappropriate to the ends aimed at.
I expect that few Spanish speakers in America would disagree that English is extremely useful, but there is a reason that not all of them have learned English, and that is because the cost was subjectively too high for them or for those who would provide language services to them. To say that we should all speak English is a little like saying we should all drive a Prius. To ignore the existence of scarcity and to ignore the present state of capital goods is to blind oneself with a utopian vision of how the economy of language ought to be arranged. That which is not seen is the real cost in capital goods that official-English policies require to bring about.
Another unseen effect of promoting English as the language of success is that often immigrant children go to school with a poor grasp of English but are forbidden or simply unsupported in using their heritage language as a means for acquiring English, and so proceed to fall behind in both languages. As Ana Celia Zentella details in her book Growing Up Bilingual, this is precisely the case for Puerto Rican immigrants in New York City, who because their parents were told by school administrators not to speak Spanish to their kids at home, grew up unable to socialize well in either English or Spanish, and are now at a distinct social and economic disadvantage in both.13 Today, this segment of the immigrant population remains statistically among the most impoverished in the country, for largely this reason. Yet no one predicted this pernicious and unseen effect of such language policies.
The only thing that is accomplished by making English the official language is a complete disenfranchisement not just of immigrant populations but of anybody who doesn't speak the particular dialect of English arbitrarily chosen as the standard. Such policies divide society into two classes of people: those who have access to the kind of upper-tier education available to the ruling elite in Washington, and everybody else. The only way to enter into this upper echelon of society is to learn this special register of English, which practically nobody grows up speaking; yet the only way to learn this particular set of language practices is to be a member of the elite in the first place. Many conservatives cry foul when they see President Obama engaged in class warfare, but they fail to recognize the same tendency within themselves, realized through more subtle means.
Such are the insidious effects of official-English policy, and yet the majority of its advocates have ignored Bastiat's and Hazlitt's admonitions. Moreover, even by their own stated goals — the pronounced goal of providing greater economic opportunity and cultural integration to immigrants — it has been shown that official-English policies in fact lead to quite opposite effects.
There is one final argument that must be made, now that our value-free praxeological analysis is complete, and that is the argument from natural rights. Official-English policies constitute nothing more than a blatant intrusion on people's natural right to interact with others in the manner and language they please, and to not have their modes of interaction dictated by the state. Just as you shouldn't be forced to interact with me in Spanish or Ojibwe, I shouldn't be forced to interact with you in English. Incentives and a free market will determine what language we ultimately converse in, whether we converse at all, or whether conversation is even necessary for our exchange. It's difficult to see how declaring English as an official language and mandating its use is conceptually any different from legislating that all my transactions must be conducted in fiat money instead of gold, or that all businesses must check for a national ID card before hiring or contracting with someone.
If English really is the language of success, the language of free enterprise, of global markets and job opportunities, as is so frequently touted, then it's hardly necessary to force anybody to learn it. As long as the market incentive is there, people will learn. But if we insist on this idea of linguistic nationalism — this socialist vision of one language, one country, one people — the result will be nothing short of English for all, and freedom for none.
- 1. Henry Hazlitt, Economics in One Lesson (Auburn: Ludwig von Mises Institute, 2008), p. 5.
- 2. Ludwig von Mises, Human Action, Pocket. (Auburn: Ludwig von Mises Institute, 2010), p. 21.
- 3. Benjamin Franklin, The Writings of Benjamin Franklin, Vol. 2: Philadelphia, quoted in Robert Lane Greene, You Are What You Speak: Grammar Grouches, Language Laws, and the Politics of Identity (New York: Delacorte Press, 2011), 227.
- 4. Indian Peace Commission, "Report to the President by the Indian Peace Commission" (Washington, D.C., 1868). Mike Reid writes about the Canadian version of these schools in "One Race to School Them All," Mises Daily, September 7, 2012.
- 5. Theodore Roosevelt, Works. Vol. XXIV, Memorial ed. (New York: Charles Scribner's 11 Sons), p. 554.
- 6. Alejandro Portes and Lingxin Hao, "E pluribus unum: Bilingualism and loss of language in the second generation," Sociology of Education 71 (1998).
- 7. Greene, You Are What You Speak: Grammar Grouches, Language Laws, and the Politics of Identity, p. 234.
- 8. It is important to note that "subjective" is not the same as "arbitrary." A person makes their decisions based on rational deliberation and individual preferences. The fact that the cost of an action has its origins in an individual's preferences is what makes the notion of efficiency "subjective." For more on the precise sense and use of subjectivity, see Ludwig von Mises, Human Action, chapter 1, section 4, "Rationality and Irrationality; Subjectivism and Objectivity of Praxeological Research," pp. 19–22.
- 9. My thanks to Jacob Hill for elucidating this point for me.
- 10. Mises, Human Action, pp. 41–44.
- 11. Jeffrey A. Tucker, It's a Jetsons World: Private Miracles and Public Crimes (Auburn: Ludwig von Mises Institute, 2011).
- 12. The original passage in its entirety is as follows: We may picture to ourselves the image of how things would be if, equipped with our present knowledge of natural resources, geography, technology, and hygienics, we had arranged all processes of production and manufactured all capital goods accordingly. We would have located the centers of production in other places. We would have populated the earth's surface in a different way. Some areas which are today densely inhabited and full of plants and farms would be less occupied. We would have assembled more people and more shops and farms in other areas. All establishments would be equipped with the most efficient machines and tools. Each of them would be of the size required for the most economical utilization of its capacity of production. In the world of our perfect planning there would be no technological backwardness, no unused capacity to produce, and no avoidable shipping of men or of goods. The productivity of human exertion would far surpass that prevailing in our actual, imperfect state. The writings of the socialists are full of such utopian fancies. Whether they call themselves Marxian or non-Marxian socialists, technocrats, or simply planners, they are all eager to show us how foolishly things are arranged in reality and how happily men could live if they were to invest the reformers with dictatorial powers. It is only the inadequacy of the capitalist mode of production that prevents mankind from enjoying all the amenities which could be produced under the contemporary state of technological knowledge. The fundamental error involved in this rationalistic romanticism is the misconception of the character of the capital goods available and of their scarcity. The intermediary products available today were manufactured in the past by our ancestors and by ourselves. The plans which guided their production were an outgrowth of the then prevailing ideas concerning ends and technological procedures. If we consider aiming at different ends and choosing different methods of production, we are faced with an alternative. We must either leave unused a great part of the capital goods available and start afresh producing modern equipment, or we must adjust our production processes as far as possible to the specific character of the capital goods available. The choice rests, as it always does in the market economy, with the consumers. Their conduct in buying or not buying settles the issue. In choosing between old tenements and new ones equipped with all the gadgets of comfort, between railroad and motorcar, between gas and electric light, between cotton and rayon goods, between silk and nylon hosiery, they implicitly choose between a continued employment of previously accumulated capital goods and their scrapping. When an old building which could still be inhabited for years is not prematurely demolished and replaced by a modern house because the tenants are not prepared to pay higher rents and prefer to satisfy other wants instead of living in more comfortable homes, it is obvious how present consumption is influenced by conditions of the past. The fact that not every technological improvement is instantly applied in the whole field is not more conspicuous than the fact that not everybody throws away his old car or his old clothes as soon as a better car is on the market or new patterns become fashionable. In all such things people are motivated by the scarcity of goods available. (Human Action, pp. 505–507)
- 13. Ana Celia Zentella, Growing Up Bilingual (Wiley-Blackwell, 1997).