These accusations are not correct. They are, instead, distortions that survive only because of the tendency of some journalists to read only what other journalists write (a phenomenon known as "pack journalism"; Parenti, 1993). When one looks at actual research, published in respectable academic journals, the picture is very different. The contrast between media reports and academic reports has been confirmed by McQuillan and Tse (1996), who reported that 87% of academic publications on bilingual education between 1984 and 1994 had conclusions favorable to bilingual education. During this same time span, media reports were only 45% favorable. 

I review here what academic research says, focusing on the impact of bilingual education on English language development and on the retention and loss of "heritage languages." 

Bilingual Education and English Language Development 
Before looking at the research, it will be helpful to first discuss how educating children in their first language can help their acquisition of another language. It seems counterintuitive to some people: If we want children to acquire English, why not teach them English? 

But using and developing the first language can help second language development a great deal. This happens in two ways. When we use the first language to teach subject matter, we give children knowledge, and this knowledge helps make the English children hear and read more comprehensible. A limited-English proficient child who knows her math, for example, thanks to math instruction in her primary language, will understand more in an English-language medium math class than a child without a good background in math. This results in better achievement in math and more English language development. 

The second way first language development helps occurs when children develop literacy in their primary language. Literacy developed in the primary language transfers to the second language. The reason literacy transfers is simple: Because we learn to read by reading, by making sense of what is on the page (Smith, 1994), it is easier to learn to read in a language we understand. Once we can read in one language, we can read in general. 

Subject matter knowledge and literacy, gained through the primary language, provide indirect but powerful support for English language development and are two of the three components of quality bilingual programs. The third component is direct support for English language development, through English as a second language classes and sheltered subject matter teaching, classes in which intermediate level ESL students learn subject matter taught in English in a comprehensible way (Escamilla, 1994; Krashen, 1996). 

What the research shows
A number of studies have shown that bilingual education is effective, with children in well-designed programs acquiring academic English as well and often better than children in all-English programs (Willig, 1985; Cummins, 1989; Krashen, 1996; Greene, 1997). Willig concluded that the better the experimental design, the more positive were the effects of bilingual education. My conclusion is that when programs have the three components described above (subject matter teaching in the first language, literacy development in the first language, comprehensible input in English), they succeed especially well (Krashen, 1996). 

The evidence used against bilingual education is not convincing. One major problem is labeling. Several critics, for example, have claimed that "English immersion" programs in El Paso and McAllen, Texas, were superior to bilingual education (e.g., Rossell and Baker, 1996). In each case, however, programs labeled "immersion" were really bilingual education, with a substantial part of the day taught in the primary language. In another study, Gersten (1985) claimed that all-English immersion was better than bilingual education. However, the sample size was very small and the duration of the study was short; also, no description of "bilingual education" was provided. For detailed discussion, see Krashen (1996). 

This framework helps answer one of the most frequently stated arguments against bilingual education: How did some immigrants do well in school without it? Here is one case, one of the many that has been described in the professional literature (Krashen, 1996, 1999, Ramos and Krashen, 1997 ; Tse, 1997). It is particularly interesting because it was published by US English as an argument against bilingual education: 

Fernando de la Pena grew up in Mexico and came to the US at age nine, with no knowledge of English. He reports that he learned English quickly, and "by the end of my first school year, I was among the top students" (de la Pena, 1991, p. 19). But de la Pena had de facto bilingual education: Had he stayed in Mexico, he would have been in the fifth grade, but when he came to the US, he was put in grade three! His knowledge of subject matter was superior to the other children in the class and he was already literate in Spanish, thanks to his education in Mexico. This helped make the input he heard comprehensible and provided a shortcut to English literacy. Cases like these provide strong support for the principles underlying bilingual education and are confirmed by numerous empirical studies showing that those who have a better education in their primary language excel in English language development (research reviewed in Krashen, 1996). 

Recent evidence against bilingual education 
Some media reports have given the impression that California's Proposition 227 was successful, that children are doing better under all-English programs than they were under bilingual education. I discuss two such reports here. 

Did LA students "take to immersion"?
Anyone glancing at the headline and opening paragraph of an article appearing in the Los Angeles Times on January 13, 1999 would get the impression that Proposition 227 was a clear success. The headline proclaimed: "L.A. students take to English immersion" and the first paragraph stated that "teachers are delivering promised reports that their children are learning English more quickly than anticipated." 

The rest of the article had a different tone. The reporter conducted (only!)13 interviews in the Los Angeles Unified School District, and concluded that children were picking up "verbal English at a surprising rate," but also reported that there were concerns that children were falling behind in their studies; many teachers were questioning "whether most of the youngsters have acquired the language skills necessary to comprehend math, reading or history lessons in English." One teacher noted that children were picking up "social English," not academic English, that new concepts still had to be presented in the primary language, and that "we won't have as many readers in our class as we did last year" (under bilingual education). Other teachers said that they had to "water down" core subjects. 

This is just what one would expect would happen. Children will pick up conversational language with any kind of program. (No comparison was made with conversational English spoken by children in bilingual programs.) The challenge is to help them develop what Cummins (1989) calls "academic language," the language of school. There were problems in this domain. Apparently, the headline writer did not read this far into the article. 

The SAT9 scores
Newspaper articles reported that LEP children in California in certain districts dramatically increased their scores on a standardized test, the SAT9, after Proposition 227 was implemented. Much of the attention was focused on Oceanside, a district that claimed to have dropped bilingual education completely. But a look at the actual scores shows that not much happened that was noteworthy. In table 1, I present SAT9 scores for all LEP children in California for 1998 and 1999, as well as scores for Oceanside. This table reveals two facts of interest: First, Oceanside's SAT9 scores for both years were very low, compared to state averages. Second, the "dramatic" increase was seen only in grade 2. In other grades, and in California in general, differences between 1998 and 1999 were quite small. And there are other questions and concerns: We have no idea what kind of a bilingual education program they had, e.g., whether it was set up in agreement with the principles outlined above. In addition, Hakuta (1999) reported that some districts that claimed growth in the SAT9 did not have bilingual education in 1998 (e.g., Westminster and Cypress), and growth was also seen in districts that kept bilingual education (Vista, Santa Ana, Ocean View). 

Table 1. SAT 9 scores for LEP children in California and Oceanside School District
 

	State of California: LEP		Oceanside: LEP
Grade	1998	1999	1998	1999
2	19	23	12	23
3	14	18	9	12
4	15	17	8	10
5	14	16	6	9
6	16	18	9	9
7	12	14	4	5
8	15	17	9	8

In my view, examining SAT9 scores is an awkward way, at best, to do research. SAT9 comparisons are very crude – one has no assurance that groups were comparable at the beginning of the year. Last year's scores do not tell us this: Among LEP children, those who acquire enough English are recategorized and are no longer LEP the next year. Also, districts differ a great deal in factors that may affect test scores, including whether and how bilingual education is done. Serious research done in a scientifically respectable way (controlled studies) consistently shows that children in quality bilingual programs outperform comparison children in all-second language classes on tests of second language literacy. The results of this kind of research are much more compelling. 

Unfortunately, these are not isolated examples. Every case reported so far of the alleged success of "immersion" in California is seriously flawed (Krashen, 1999; McQuillan, 1998a; Krashen and McQuillan, 1999). 

Improving bilingual education 
Bilingual education has done well, but it can do much better. The biggest problem, in my view, is the absence of books, in both the first and second language, in the lives of students in these programs. It is now firmly established that reading for meaning, especially free voluntary reading, is the major source of our literacy competence. Those who report that they read more read better and write better (Krashen, 1993), and students who participate in free reading activities in school (e.g., sustained silent reading) show superior literacy development when compared to students who do not (Krashen, 1993; Elley, 1998). Free reading appears to work for first language, for second language, for children, and for teenagers, and the research has confirmed this in many different countries. Free voluntary reading can help all components of bilingual education: It is a source of comprehensible input in English, a means for developing knowledge and literacy in the first language, and, as we will see later, a way of continuing first language development. 

It is also firmly established that those with greater access to books read more; while access is not sufficient to guarantee reading, it is certainly necessary (Krashen, 1993; McQuillan, 1998b). It is also very clear that many limited English proficient children have little access to books in any language. I present here data on Spanish-speaking children. 

The average Hispanic family with limited English proficient children has about 26 books in their home (Ramirez, Yuen, Ramey, and Pasta, 1991). This refers to the total number of books in the home, including the bible, cookbooks, and dictionaries. This is about one-sixth the US average (Purves and Elley, 1992). School is not helping: In fact, school is making things worse. Pucci (1994) investigated school libraries in schools with strong bilingual programs in Southern California and found that books in Spanish were very scarce. Those that were available, while often of high quality, were usually short and for younger children. 

Enriching the print environment is not the only recommendation one can make in discussing improvement of bilingual education, but it is a great place to begin. If it is true that learning to read in the primary language is in fact beneficial, children need something to read. My suggestion is a massive book flood in the child's home language as well as in English, a suggestion that is relatively inexpensive to implement. 

The Retention and Loss of Heritage Languages: Are Immigrants Resisting English?
Language shift: A powerful force
One of the most consistent findings in the field of sociology of language is the phenomenon of language shift: Heritage languages are usually not maintained and are rarely developed. This fact is nearly unknown to the general public, as well as to many politicians. Robert Dole, for example, felt that immigrants were resisting English, and maintained that we need "the glue of language to help hold us together" (quoted in the Los Angeles Times, October 31, 1995). Newt Gingrich also warned that "Immigrants need to make a sharp break with the past ..." (Los Angeles Times). 

Here are just a few of the many studies showing that "shift happens": Hudson-Edwards and Bills (1980) examined self-report of ability in Spanish among residents of a section of Albuquerque considered to be a strong Spanish-speaking community. As seen in table 2, the older generation considered themselves to be better in Spanish than English, but their children rated themselves more highly in English. 

Table 2. Self-report of ability in Spanish and English (percent claiming "good" or "very good" ability) 
 

Senior = heads of households, spouses, siblings; Junior = children of heads of households. Source: Hudson-Edwards & Bills, 1980, Albuquerque.

Table 3. Self-reported competence in English and in parents' language 
 

	Knows English		Knows parents' language		Prefers
	well	very well	well	very well	English
Total	93.6	64.1	44.3	16.1	72.3
Mexican	86.1	43.7	69.1	34.9	44.8

Source: Portes and Hao (1998) 

Orellana, Ek and Hernandez (1999) conducted conversations and interviews with Mexican-American children in bilingual schools in Los Angeles, and observed "a gradual but marked shift over the middle childhood years toward a preference for English, and a disinclination to use Spanish. When we spoke in English at the start of the year in (a) first-grade classroom, the children called out for Spanish. When we spoke in Spanish in the focus groups with fifth graders, all but the children who arrived in the U.S. within the last year responded in English, and several complained, saying 'Aw, do we have to speak Spanish?' ...." (pp. 125-26). 

Why does shift occur?
The most obvious cause of shift is lack of input in the heritage language. Input/use related variables are clear predictors of heritage language competence. 

Some of these input factors may be beyond the control of the subject. A number of studies have confirmed that heritage language competence is related to parental use of the heritage language (HL) (Portes and Lao, 1998; Hinton, 1999; Kondo, 1998; Cho and Krashen, 2000). Parental use, however, appears to be necessary but not sufficient. Hinton (1999) reported that in her sample, "many of the families ... did in fact choose to use the heritage language at home, and yet still found that their children were loosing fluency" (see also Kondo, 1998). Not surprisingly, studies also show that those who live in close proximity to other HL speakers maintain it longer (Demos, 1988), an effect that appears to be especially predictive of HL maintenance after the first generation (Li, 1982). Of course, once the speaker moves away from other HL speakers, competence may diminish (Hinton, 1999). Also, those who visit the country of origin more often have higher HL competence (Demos, 1988; Kondo, 1988, Hinton, 1999; Cho and Krashen, 2000). Other input factors, such as reading and watching TV (Cho and Krashen, 2000), are under the voluntary control of the HL speaker. 

Less obvious are affective factors, but they appear to be quite powerful. Tse (1998a) notes that some some language minority group members go through a stage in which the desire to integrate into the target culture is so strong that there is apathy toward or even rejection of the heritage culture. Tse refers to this stage as Ethnic Ambivalence or Ethnic Evasion. Typically, this stage occurs during childhood and adolescence, and may extend into adulthood. Those in this stage have little interest in the heritage language, and may even avoid using it. 

"Maria Shao recounted how her knowledge of Chinese was a source of shame. She recalled that when she was in elementary school, 'if I had friends over, I purposely spoke English to my parents. Normally, we only spoke Chinese at home. Because of the presence of a non-Chinese, I used to purposely speak English.'" (Tse, 1998, p. 21). 

"David Mura noted these feelings as a child: 'I certainly didn't want to be thought of as Japanese-American. I was American, pure and simple. I was proud I didn't know Japanese, that English was my sole tongue.'" (p. 21) 

For some, this stage gives way to another stage, Ethnic Emergence, in which minority group members get interested in their ethnic heritage. Those in this stage, Tse points out, may be quite motivated to develop their competence in the heritage language. 

Another affective factor is a reluctance to use the language because of the negative reactions of other HL speakers. Some imperfect HL speakers (often a younger sibling) report that their efforts to speak the heritage language are met with correction and even ridicule by more competent HL speakers, a reaction that discourages the use of the HL, and thus results in less input and even less competence. What is often lacking are late-acquired aspects of language, aspects that typically do not interfere with communication but that indicate politeness or mark social class differences. 

In Krashen (1998a) I presented some cases of "language shyness." Subjects confirmed that correction and ridicule discouraged their use of the heritage language. Here is one example: 

"I began to realize as I spoke Spanish to my relatives, they would constantly correct my grammar or pronunciation. Of course, since I was a fairly young child the mistakes I made were 'cute' to them and they would giggle and correct me. This ... would annoy me to no end. I wasn't trying to be 'cute'; I was trying to be serious. My relatives would say, 'You would never know that you are the daughter of an Argentine.' Comments like these along with others are what I now believe shut me off to Spanish ....". 

"My self-esteem reached an all-time low in college. Several of my peers made well-meaning, but harsh comments upon hearing my Spanish. This was the final blow. It was then I made the decision that I wouldn't speak unless I could speak fluently, grammatically correct, and with a proper native accent. I couldn't even feel comfortable describing myself as bilingual on my resume. I had to add 'limited proficiency' in parentheses to ease my conscience ... I was ashamed of being Puerto Rican and living in a bilingual home and never learning Spanish ... the only conclusion I could come to was that it was somehow my fault ...". 

Developing the heritage language
If it is worthwhile to develop the HL, how can it be done? The usual solution is formal language classes, either those meant for non-native speakers or specially designed classes ("Spanish for Native Speakers"). 

Heritage language speakers are in a no-win situation in foreign language classes. If they do well, it is expected. If HL speakers do not do well in foreign language classes, the experience is especially painful. Often, classes focus on conscious learning of grammatical rules that are late acquired. Some HL speakers may not have learned or acquired these items. It can happen that non-speakers of the HL who are good at grammar will outperform HL speakers on grammar tests and get higher grades in the language class, even though the non-speaker of the HL may be incapable of communicating the simplest idea in the language, while the HL speaker may be quite competent in everyday conversation. Such events could be psychologically devastating, a message to the HL speaker that he or she does not know his or her own language, while an outsider does. Even though the kind of knowledge the outsider has is not genuine, the HL speaker may not understand this, given the authority of the classroom and the value the teacher places on conscious knowledge of grammar. 

Some heritage language programs have been successful, particularly those that are integrated into the school day (Tse 1998b). McQuillan (1998c) describes two heritage language classes for university students (Spanish for native speakers) that not only succeeded, but provided a foundation for future progress. Both classes included a survey of popular literature as well as self-selected reading. Students showed clear gains in language (vocabulary) and, more importantly, when students in one class were surveyed seven months after the class ended, they were reading more in Spanish on their own than a comparison group. 

McQuillan's results strongly suggest that providing a print-rich environment is also a strong investment in heritage language development. If heritage language speakers become readers in their primary language, they can continue to develop their primary language, whether or not other sources of input are available. Reading is also the perfect method for heritage language speakers who do not want to risk errors in interacting with others: It is the perfect method for the shy language acquirer. 

If immigrants are dropping their heritage language and embracing English, why do we need bilingual education? When immigrants acquire English informally, the version they acquire is what Cummins (1989) terms "conversational language," the language of everyday interaction. They do not necessarily acquire "academic language," the language of school. Evidence for this is the Los Angeles Times report on the "success" of 227, as reported earlier. Evidence also includes studies such as Romo and Falbo (1996)' s investigation of 100 Latino high school students designated as being at risk for dropping out. Romo and Falbo reported that "almost all students in our sample were comfortable speaking in English ... yet, almost all students in our sample experienced a skills deficit in reading" (p. 9); although the students were in the seventh to eleventh grades, their average reading score was sixth-grade. In other words, they had acquired conversational, but not academic English. 

As noted earlier, good bilingual education programs aid in the development of academic English by providing literacy in the first language, which transfers to English, subject matter teaching in the primary language, which provides background knowledge that makes English input more comprehensible, as well as comprehensible subject matter teaching in English. The arguments presented in the second half of this paper indicate that an additional component would be desirable: Continuing development of the heritage language. 

References
Cho, Grace, and Krashen, Stephen. 1998. The negative consequences of heritage language loss and why we should care. In Heritage Language Development, ed. Stephen Krashen, Lucy Tse, and Jeff McQuillan, 31-39. Culver City, CA: Language Education Associates. 

Cho, Grace, and Krashen, Stephen. 2000. The role of voluntary factors in heritage language development: How speakers can develop the heritage language on their own. ITL: Review of Applied Linguistics. 127-128: 127-140. 

Cho, Grace., Cho, Kyung-Sook and Tse, Lucy. 1997. Why ethnic minorities want to develop their heritage language: The case of Korean-Americans. Language, Culture and Curriculum 10 (2): 106-112. 

Cummins, Jim. 1989. Empowering Minority Students. Ontario, CA: California Association for Bilingual Education. 

de la Pena, Fernando. 1991. Democracy or Babel? The Case for Official English in the United States. Washington, DC: US English. 

Demos, Vasikilie. 1988. Ethnic mother-tongue maintenance among Greek Orthodox Americans. International Journal of the Sociology of Language 69: 59-71. 

Elley, Warwick. 1998. Raising Literacy Levels in Third World Countries: A Method that Works. Culver City, CA: Language Education Associates. 

Escamilla, Kathy. 1994. The sociolinguistic environment of a bilingual school: A case study introduction. Bilingual Research Journal 18:1,2: 1-29. 

Fishman, Joshua. 1990. Empirical explorations of two popular assumptions: Inter-polity perspective on the relationships between linguistic heterogeneity, civil strife, and per capita gross national product. In Learning in Two Languages, ed. Gary Imhoff, 209-225. New Brunswick, NJ: Transaction Publishers.

Gersten, Russell. 1985. Structured immersion for language minority students: Results of a longitudinal evaluation. Educational Evaluation and Policy Analysis 7: 187-196.

Greene, Jay. 1997. A Meta-Analysis of the Rossell and Baker review of bilingual education research. Bilingual Research Journal 21(2,3): 103-122. 

Hakuta, Kenji. 1986. Mirror of Language: The Debate on Bilingualism. New York: Basic Books. 

Hakuta, Kenji. 1999. SAT-9 scores and California's Proposition 227. NABE News 22 (8):1, 6-7. 

Hinton, Leanne. 1998. Involuntary language loss among immigrants: Asian-American linguistic autobiographies. Paper presented at Georgetown Round Table for Languages and Linguistics, May 1998. 

Hudson-Edwards, Alan. and Bills, Garland. 1980. Intergenerational language shift in an Albuquerque barrio. In A Festschrift for Jacob Ornstein, ed. Edward Blansitt and Richard Tescher, 139-158. New York: Newbury House. 

Kondo, Kimi. 1998. Social-psychological factors affecting language maintenance: Interviews with Shin Nisei university students in Hawaii. Linguistics and Education 9 (4): 369-408. 

Krashen, Stephen. 1993. The Power of Reading. Englewood, CO: Libraries Unlimited. 

Krashen, Stephen. 1996. Under Attack: The Case Against Bilingual Education. Culver City, CA: Language Education Associates. 

Krashen, Stephen. 1998a. Heritage language development: Some practical arguments.In Heritage Language Development, ed Stephen Krashen, Lucy Tse, and Jeff McQuillan, 3-13. Culver City, CA: Language Education Associates. 

Krashen, Stephen. 1998b.Language shyness and heritage language development. In Heritage Language Development, ed. Stephen Krashen, Lucy Tse, and Jeff McQuillan, 41-49. Culver City, CA: Language Education Associates.

Krashen, Stephen. 1999a. Condemned Without a Trial: Bogus Arguments Against Bilingual Education. Portsmouth, NH: Heinemann. 

Krashen, Stephen. 1999b. Bilingual education: Arguments for and (bogus) arguments against. Paper presented at the Georgetown Round Table on Languages and Linguistics, Georgetown University, Washington, D.C., May 1999. 

Krashen, Stephen. and McQuillan, Jeff. 1999. Structured immersion falls short of expectations: An analysis of Clark (1999). NABE News 23(1): 13-15. 

Li, Wen Lang. 1982. The language shift of Chinese-Americans. International Journal of the Sociology of Language 38: 109-124. 

McQuillan, Jeff. 1998a. Is 99% failure a "success?" Orange Unified's English immersion program. The Multilingual Educator 21 (7):11. 

McQuillan, Jeff. 1998b. The Literacy Crisis: False Claims and Real Solutions. Portsmouth, NH: Heinemann.

McQuillan, Jeff. 1998c. The use of self-selected and free voluntary reading in heritage language programs: A review of research. In Heritage Language Development, ed. Stephen Krashen, Lucy Tse, and Jeff McQuillan, 73-87. Culver City, CA: Language Education Associates. 

McQuillan, Jeff, and Tse, Lucy. 1996. Does research really matter? An analysis of media opinion on bilingual education, 1984-1994. Bilingual Research Journal 20 (1): 1-27. 

Orellana, Marjorie Faulstich, Ek, Lucilia, and Hernandez, Arcelia. 1999. Bilingual education in an immigrant community: Proposition 227 in California. International Journal of Bilingual Education and Bilingualism 2 (2): 114-130. 

Parenti, Michael. 1993. Inventing Reality: The Politics of the News Media. New York: St. Martins Press. 

Portes, Alejandro. and Hao, Lingzin. 1998. E Pluribus Unum: Bilingual and loss of language in the second generation. Sociology of Education 71: 269-294. 

Pucci, Sandra. 1994. Supporting Spanish language literacy: Latino children and free reading resources in schools. Bilingual Research Journal 18 (1-2): 67-82.

Purves, Alan. and Elley, Warwick. 1994. The role of the home and student differences in reading performance. In The IEA Study of Reading Achievement and Instruction in Thirty-Two School Systems, ed. Warwick Elley, 89-121.Oxford: Pergamon. 

Ramirez, J. David., Yuen, Sandra., Ramey, Dena, and Pasta, David. 1991. Final Report: Longitudinal Study of an English Immersion Strategy and an Early-Exit Transitional Bilingual Education Program for Language-Minority Children. Volume I. San Mateo, CA: Aguirre International. 

Ramos, Francisco and Krashen, Stephen. 1997. Success without bilingual education? Some European cases of de facto bilingual education. CABE Newsletter 20,6:7,19. 

Romo, Harriett and Falbo, Toni. 1996. Latino High School Graduation: Defying the Odds. Austin: University of Texas Press. 

Rossell, Christine and Baker, Keith. 1996. The educational effectiveness of bilingual education. Research in the Teaching of English 30: 7-74. 

Smith, Frank. 1994. Understanding Reading. 5th Ed. Hillsdale, NJ: Erlbaum.

Tse, Lucy. 1997. A bilingual helping hand. Los Angeles Times, Dec.17. 

Tse, Lucy. 1998a. Ethnic identify formation and its implications for heritage language development. In Heritage Language Development, ed. Stephen Krashen, Lucy Tse, and Jeff McQuillan, 15-27. Culver City, CA: Language Education Associates. 

Tse, Lucy. 1998b. Affecting affect: The impact of heritage language programs on students attitudes. In Heritage Language Development, ed. Stephen Krashen, Lucy Tse, and Jeff McQuillan, 51-72. Culver City, CA: Language Education Associates. 

Willig, Ann. 1985. A meta-analysis of selected studies on the effectiveness of bilingual education. Review of Educational Research 55: 269-316. 

Wong Fillmore, Lilly. 1991. When learning a second language means losing the first. Early Childhood Research Quarterly 6: 323-346.