Every month, the Harvard Graduate School of Education invites educators, researchers, community activists, and policymakers from across the country to talk about key issues in schools and school reform. We are pleased to be able to provide you with an edited transcript of some of these forums. Below is an edited transcript of a talk given at the Harvard Graduate School of Education, October 15, 2001.
For easier reading, we have divided the transcript into the following sections:
You can also scroll right through the transcript without clicking on the above links.
DOTTIE ENGLER: We’re delighted to have with us Ron Unz, who is a theoretical physicist. This is actually a return for him, as he graduated from Harvard as an undergraduate. Mr. Unz is the chairman of Wall Street Analytics, a financial services software company that he founded in 1987. In addition to his degree from Harvard, he holds graduate degrees from Cambridge University and Stanford. He has long been interested in public policy issues. He has authored hundreds of articles that have appeared on issues ranging from campaign reform to school vouchers to China appearing in publications like the Wall Street Journal, the New York Times, National Review, and The New Republic.
In 1994, Mr. Unz launched a surprise Republican primary challenge to California’s incumbent governor Pete Wilson, running on a conservative, pro-immigrant platform against the prevailing political sentiment, and received 34 percent of the vote, which is extraordinary for a novice. Later that year he campaigned as a leading opponent of Proposition 187, the anti-immigration initiative, and spoke at many rallies, including one attended by 70,000 people. It was part of a pro-immigrant march in Los Angeles and it was the largest political rally in California history. So you can see that it is very hard to typecast him in terms of his political persuasion, and I think he loves to keep it that way.
In 1997, Mr. Unz began the English for the Children initiative campaign in California after learning of boycotts by Hispanic parents against Spanish-language programs in the Los Angeles area. He drafted Proposition 227 and led the campaign to qualify and pass the measure, culminating in a victory at the polls on June 2, 1998.
Following the passage of the California measure, he assisted activists in Arizona in passing their own anti-bilingual education initiative. He has now established a national advocacy organization, English for the Children, to replace bilingual education with English immersion throughout the country. Today it was announced that enough signatures had been secured to get the bilingual initiative on the ballot in Massachusetts in November.
Debating him this evening will be Catherine Snow, the Henry Lee Shattuck Professor of Education here at the Harvard Graduate School of Education. Catherine is an expert on language and literacy development in children, focusing on how oral language skills are acquired and how they relate to literacy outcomes. She recently chaired a National Academy of Sciences committee that prepared the report, “Preventing Reading Difficulties in Young Children.”
Her current research activities include a longitudinal study of language and literacy skills among low-income children who have been followed for thirteen years since age three; various attempts to develop consensus among teacher educators concerning what pre- and in-service reading teachers need to know about language and about literacy; and research following the language development of young children participating in early Head Start intervention as well as the vocabulary development of first- and second-language learners.
She’s also considering various aspects of transfer from first to second language in the domains of language and literacy. Catherine has written about bilingualism and its relation to language policy issues such as bilingual education in the United States and in developing nations, and about testing policy.
From 2000-2001, she was president of the American Educational Research Association. She has written a number of books and publications including Starting Out Right: A Guide to Promoting Children’s Reading Success, Preventing Reading Difficulties in Young Children, and an article for Psycholinguistics entitled, “Bilingualism and Second Language Acquisition.”
Moderating is Toni Randolph of WBUR [public radio], where she has worked for five and half years. She began her broadcast journalism career in commercial radio, a little bit of a dirty word in public radio, but I guess they’ve forgiven you. And soon after graduating from the State University of New York at Buffalo with a degree in journalism she also earned a master’s degree from the Graduate School of Journalism at Columbia.
So from now on, Toni is ruling the roost. She may look gentle, but I can assure you she’s quite firm. So thank you all for coming tonight, and we’re off!
TONI RANDOLPH: Thank you, Dottie. Good evening, everyone, and thank you very much for coming. Before we hear from tonight’s panelists, I just want to give you a few facts about bilingual education. About five percent of the Massachusetts’s 976,000 students are what is defined as LEP or Limited English Proficient. Basically that means that English is not their first language. There are about 69 different languages spoken in Massachusetts schools. After English, Spanish is the most widespread, but other languages include Portuguese, Chinese, Vietnamese, Khmer, Korean, Russian, French, Creole, Greek, and Arabic, to name just a few.
Under current Massachusetts law, a district must provide bilingual educational classes when at least 20 of its students have the same native language. Students can take classes in their native language for up to three years before they’re moved into regular education classrooms.
All other students with limited English skills are taking English as a Second Language classes, which are taught only in English. There are several efforts to change bilingual education in Massachusetts, and under the proposal supported by Mr. Unz, which we’ll be discussing tonight, students whose first language is something other that English would be placed in intense English classes for a year before they’re placed in regular education.
Mr. Unz, along with many Massachusetts residents, some of whom may be in this hall tonight, have been collecting signatures and apparently have enough to place the measure on the ballot for 2002. Voters will decide whether to change bilingual education, [something Mr. Unz has achieved] successfully in California and Arizona.
The format for tonight’s debate will allow each speaker up to eight minutes to give opening remarks. Then each will have three minutes of rebuttal time to respond to the others’ statements. After that, I’ll have a few questions for each of them, and then we’ll open it up to the audience. So now it’s time to hear from our panelists, and we’ll start with Mr. Unz.
RON UNZ: Thank you very much, and thanks for the opportunity to speak before you this evening. One question that came up during the California campaign is how in the world did I get involved in this issue? In other words, it seems I speak English perfectly well. Why would I have got involved in this issue to begin with?
The answer is that I actually come from a little bit of an immigrant background myself in that my mother was born in Los Angeles, but she grew up not speaking a word of English. Then when she was a young child, about four or five years old, she learned English very quickly and easily. For that reason, when I’d first heard about these bilingual programs back when I was in junior high school, they never really seemed to make any sense to me. Why didn’t the schools simply teach English to these children as soon as they began school rather than keeping them in these other programs sometimes for many years? According to all the news reports I had been reading, those programs didn’t really seem to work very well. When something doesn’t seem to make much sense and it doesn’t seem to work very well, you assume it probably will gradually go away. But instead, over the 20- or 25-year period, bilingual education grew dramatically.
Then in 1996 I read a series of articles in the Los Angeles Times about a group of immigrant Latino parents in downtown Los Angeles, very poor garment workers, who began a public boycott of their own local elementary school because it refused to teach the children English. Parents had to carry picket signs outside a public school because it refused to allow their children to learn English. I felt something finally had to be done about it.
When I began researching the issue of bilingual education as practiced in California, I found some absolutely horrifying statistics. The official data of California at that time, coming from the State Office of Bilingual Instruction, showed that a quarter of all the children in California public schools didn’t know English. A quarter of 1.5 million students! Of the students who didn’t know English, each year only five or six percent allegedly learned English. In other words, 95 percent of all the students in California who started a school year classified as not knowing English ended that school year still classified as not having learned English.
Any program that seems to have an official 95 percent annual failure rate needs changing.
So I organized an effort to put a measure on the ballot to shift the state of California away from the these native language oriented so-called bilingual education programs towards a simple and effective system of intensive, sheltered English immersion.
The idea behind it is simple. When young children start school not knowing English, they would normally be placed in a special classroom with the other children who are also learning English to teach them English as quickly as possible over a period of a few months or a year or possibly even longer than a year. Once they learned English, then they would be moved into the regular classrooms with all the other students. It seemed to make a lot of sense to me, and it also made a lot of sense to the voters of California.
The campaign we had was probably the most bipartisan initiative campaign in the history of California in that it was opposed by nearly all the Democrats and all the Republicans. It was opposed by President Bill Clinton. It was opposed by all four candidates for Governor, Democrat and Republican. It was opposed by nearly every union, nearly every educational organization, nearly all the newspapers in the state, and all the political slates. We were also outspent on advertising by a ratio of 25 to 1. Nonetheless, we won on a huge landslide, one of the biggest landslides of any contested initiative campaign in the history of California.
Once the political dust had settled and our initiative had been ruled absolutely constitutional by four separate federal judges in a matter of a few weeks, then the reporters started going into the classroom at the beginning of the new school year to see the impact of this allegedly disastrous, catastrophic measure which would sweep away a program that had supposedly been so successful. The results were quite interesting.
Virtually every single article published in California in the last three or four years by newspapers that originally opposed the initiative has been extremely complimentary, almost flattering. These bilingual teachers, or rather ex-bilingual teachers and ex-bilingual administrators in many cases, are saying, “We opposed the measure every step of the way. We thought it would be a disaster. But it’s working incredibly well. The children are learning English so much more quickly than we ever imagined.”
Then the test scores started coming in. Now, the proof of the pudding of any public policy measure is in the eating. Test scores are a way to see whether something works or something doesn’t work. And in the State of California, what we’ve seen over the last few years is perhaps the most dramatic single rise in academic performance by a large group of students-the immigrant students of California-recorded almost anywhere in the country.
What we have seen is that in less than two years after the implementation of the new initiative, the average mean percentile test scores of over a million immigrant students went up by 40 percent. And that 40 percent rise includes those school districts in the state of California that dragged their heels on implementing the initiative.
Those that most completely got rid of their bilingual programs were able to double their test scores in less than two years. In fact, this last year the test scores of California’s immigrant students again rose more than twice as quickly as the test scores of the non-immigrant students. That is the sort of thing that should be done in Massachusetts as well.
When I was laying the groundwork for the campaign, I talked with some of America’s leading advocates of bilingual education as well as many of the critics. Virtually none of those who supported bilingual education-and that includes organizations, academics, activists, individuals-would defend the program as it existed, as it was practiced. They all made excuses for the fact that the program was such a dismal failure. They would say, “The program admittedly works very badly in these different states, in these different cities. We in no way will defend this program as practiced in California.” The problem, they said, is not the theory but the practice. The program does not have enough teachers, does not have the right teachers, does not have enough money, does not have the correct curriculum, does not have enough administrative support.
So I asked them then, “Can you point to any large-scale example anywhere in the United States where bilingual education has actually worked well?” They couldn’t point to a single one. Now, I’m a theoretical physicist by training. In science, there is a huge difference between theory and experiment. If you have a theory, even one backed by leading Harvard University professors, that you’ve tried over and over and over again throughout the entire United States for a period of 30 years, and it has never actually worked once in practice on a large scale, I say the rules of science are that you throw away the theory, you admit it is wrong, and you switch to something that does work.
In about a year’s time, the people of Massachusetts will have a chance to junk this failed theory of bilingual education which has never worked anywhere on a large scale in the United States of America and switch to something that does work, which is intensive English immersion. This will double the test scores of Massachusetts immigrant students as it is currently doing for over a million immigrant students in California. Thank you very much.
CATHERINE SNOW: Let me add my thanks to Ron Unz as to all of you for being here, and my condolences for those of you who are getting hotter and more uncomfortable in your positions. Let me also say that I admire Ron Unz. I think Ron Unz has done something that constitutes good citizenship in many ways. He has focused his attention and his remarkable energy and vigor on issues of great importance in public education. And he has made this debate possible. I just wish that the vigor with which he approached these issues were matched by a rigor in thinking about them and in informing himself about the real facts of bilingual education and of second language acquisition and bilingualism. I don’t have very much time. Let me just try to make a few points about the basic facts of bilingual development and bilingual education.
First of all, I agree with Mr. Unz. Quality of educational programs counts. In every single kind of educational policy, it doesn’t matter what the theory is if you don’t have the theory well implemented. Badly designed bilingual programs are not programs that I would defend, or anybody else would defend. On the other hand, I would point out that intensive one-year immersion programs are not particularly well designed. In fact, there is no design for one-year immersion in California. The resources that are available to support teachers trying to implement one-year immersion are, if anything, far worse than the resources available to teachers implementing bilingual education.
So the alternative here, a one-year immersion program, is a program for which poor quality is also almost inevitably the case. And it’s not the case that there are no good bilingual programs in the country. There are many. There are too many poor bilingual programs, but there are many good bilingual programs.
Think about math and science in American schools. There are a lot of classrooms in which math and science are being taught fairly badly, with poor curricula and bad teacher preparation. We don’t eliminate math and science from the curriculum because it isn’t being done well.
Secondly, Mr. Unz has publicly challenged the idea that it might take longer than a year or, as he said tonight, a year and a little bit, to learn English. And, of course, I agree with him. You can learn some English in a year. You can learn quite a bit of English in a year. How many vocabulary items do we think the children in these intensive immersion programs able to learn? Maybe 1,000? Maybe 1,200? Maybe 1,500? That would be a very good outcome at the end of one year of intensive immersion. But still it puts children several thousand words behind monolingual English speakers who have been acquiring English since birth. And we know from studies of English-learning children and English-only children that if they arrive at first grade with vocabularies smaller than 5,000 words or 6,000 words, they’re very likely to have trouble learning how to read.
Learning how to read is the big challenge in school. It’s not speaking English; it’s reading English. And children who start the process of learning how to read with limited vocabularies in English are going to have trouble. That is the situation of children who have only had one year of not very good quality intensive immersion.
Third, outcomes at the end of kindergarten or even at the end of first grade are really not the outcomes we as educators should be interested in. The long-term outcomes that are of interest to me at least are outcomes in fourth grade reading comprehension. They’re outcomes in 8th-grade math. I want to know whether children can understand stories and predict what might happen next, or interpret what the characters are doing. I want to know if they can explain their own reasoning when they do a math problem. I want to know if they can formulate their observations and test hypotheses in their science classes. I want to know whether they’re going to get into algebra and trigonometry in high school, whether they’re going to get admitted to colleges. Those are the outcomes that really count.
Now, Mr. Unz claims that the SAT-9 scores in California have gone up because of bilingual education. That’s a very dubious claim. But even if it were true that first- and second-graders’ SAT-9 scores had gone up because of the elimination of bilingual education, these are not outcomes which are of long-term great relevance for English-language learners. Let’s think about the challenges of understanding complex text in fourth grade and ask ourselves whether one year of intensive exposure to English is sufficient to prepare children for those tasks.
Finally, with reference to bilingual education, it is clear the bilingual programs do teach English. That’s why they’re called “bilingual.” The presentation that has been given here is that children are in other-language programs until they are exited from other-language programs. Of course, good bilingual programs teach English; and, of course, bilingual educators, like the parents of immigrant children, want children to learn English.
Some of them might also-the parents as well as the educators-want children to maintain a home language. And there is no reason why that needs to be in conflict with the task of learning English. So the notion that bilingual education programs are not focused on teaching English is, of course, just a simple misrepresentation of what is going on in the good programs that we would like to try to replicate.
Let me move to a slightly different level with my comments; comments not about bilingual education, which I could go on talking about a lot longer, but about making better educational policy. As I said, I really admire Mr. Unz in his commitment to improving education, and I wish there were more people out there who are citizens who are as interested.
However, responsible advocacy is informed advocacy. I wish he would pay attention to the research in this area. But even if he wants to reject educational research, he could pay attention to the linguists and the psycholinguists who are studying second-language acquisition. I don’t have time to quote chapter and verse of the research about, for example, how long it takes for children to acquire a reasonable vocabulary and control over English that will equip them to do good academic work. But there is a pile of books over there. I don’t have time to talk about them. I hope perhaps Mr. Unz will have time to read them, because they would inform him about the realities of second language acquisition.
Secondly, good educational policy is unlikely to be made by referendum. Good educational policy is unlikely to be made in such a way that there is a fiat, a single program for all children, a one-size-fits-all policy. That’s not the way good education works. Education is very complex, and silver bullets do not solve challenging problems. Of course, I should point out the even if Mr. Unz got his way and transitional bilingual education were eliminated, some negative consequences that he would acknowledge as negative consequences would come along with that. Older children whom he agrees need bilingual education probably won’t get it. Two-way programs and maintenance bilingual programs that are actually producing bilinguals much better than foreign language classrooms in U.S. schools are will also be eliminated in the process.
Finally, he suggests that parents boycotted bilingual programs because they didn’t have any alternative. The fact of the matter is that in every state in the union, parents have a choice about bilingual or English-only programs. They simply have to ask to get their children out of bilingual. We do not need to destroy bilingual education in order to protect the rights of individual parents to choose English-only programs if that’s what they want.
RON UNZ: Professor Snow raised a number of very important issues which I’m glad I’ll now be able to address. The nature of this debate is a little bit strange to me because it mimics so much the debate I had several years ago during the California campaign. The difference is that the facts are now in. The largest controlled educational experiment in the history of the world took place a few years ago involving over a million students in California who were largely shifted away from bilingual education to English immersion. During that campaign, the supporters of bilingual education refused to defend their existing program. They admitted they had terrible problems, but they said the alternative would be worse. Getting rid of these programs would be a disaster.
Instead, the average test scores of over a million immigrant students have gone up by 50 percent in less than three years. Those school districts that most strictly and completely followed the initiative and got rid of their bilingual programs doubled their test scores in less than three years. Don’t believe me. Believe The New York Times, The Washington Post, CBS News, every major media source in the United States.
The war is over, or at least it should be over if academics were willing to look at the reality of the world rather than their own research. And, again, don’t take my word for it. One of the strongest advocates we have now for Prop 227 in California is an individual who for 30 years was a bilingual education teacher. He became a school superintendent. He ended up opposing the initiative when it was on the ballot. He urged people to vote no.
When it passed, though, he said, “The law is the law. The law has to be obeyed.” He followed the initiative, got rid of his bilingual programs. His immigrant students’ test scores doubled in less than two years. And he became a born-again convert to English immersion. He now admits he was wrong for 30 years. Bilingual education does not work. English immersion does work. And he was the founder of the California Association of bilingual educators. He admits he was wrong for 30 years.
Or take for example a man named Reed Hastings. He’s the liberal Democrat who is the president of the California State Board of Education, appointed by Governor Gray Davis. He opposed the initiative when it was on the ballot; he said it was one-size-fits-all. The results have been so phenomenally good that he has been trying to get those school districts that have blocked the initiative and tried to avoid implementing it to back down and follow the law.
He originally grew up in Massachusetts, and when he found out I was doing an initiative here, he not only endorsed it but he wrote a $20,000 check to underscore his endorsement because he wants to improve the education of immigrant children in Massachusetts just the way it happened in California.
Now, again, I’m a theoretical physicist. I’m somebody who comes from a theoretical physics background. I’m an academic by training. I think academics should look at the reality of the world rather than the theories that are published in a lot of books which may or may not be correct. The point about it is that all the academics who support bilingual education, they all believe-and publicly, many times are willing to state-that it takes a young child five to seven years to learn English. Some of them actually claim ten years. I would say this: I urge every voter in Massachusetts who honestly believes it takes five to seven years for a young child to learn English to vote no on our initiative.
Make up your own mind. If you really think a child who starts kindergarten in the United States will require until junior high to actually learn English, I urge you to oppose our initiative. Furthermore, the academics who support bilingual education all claim that the older you are the easier it is to learn another language. They claim adults learn a new language easier than teenagers; teenagers easier than little children. They are the only people in the world who believe that. Anyone who honestly believes that the older you are the easier it is to learn another language should vote no on our initiative; it’s as simple as that.
For the other 99.5 percent of the people in Massachusetts or the United States, I urge you to vote yes. Again, the proof of the pudding is in the eating. The test scores of over a million immigrant students in California have risen dramatically. Their parents are happy, their teachers are happy. The people who opposed the initiative now support it.
Now, why would that be happening if I were wrong and if the initiative didn’t work? I think you have to really ask yourself whether the problem in these immigrant test scores we have seen over all these years is in spite of bilingual education or is because of bilingual education. Think of a very simple fact. In California and in the rest of the United States, children enter the public schools speaking up to 140 different languages. In almost all of these parts of the country, including California, the only group of students that receives any significant quantity of these allegedly beneficial bilingual programs is Latino students. All the other language groups are given the equivalent of English immersion immediately. They’re taught English as soon as they start school.
Now, by an amazing coincidence, the one immigrant group in the United States that does the worst in school, with the highest dropout rates and the lowest test scores and the lowest rate of admissions to college, is that of Latino students. That doesn’t prove that the fault is bilingual education. But if the one group that gets the most bilingual education of all the immigrant groups does the worst in school, it certainly starts to shift the burden of proof over to the other side.
Now that getting rid of these programs in California has so dramatically improved the academic performance of those Latino students, you really have to ask yourself whether you should believe the reality of your own senses, the test scores of a million immigrant students in California, or four or five books piled there written by some professors at Harvard and a few other universities. Again, reality trumps theory. Theory cannot defeat reality. And the reality of a million students in California is not something that a few professors can whist away by citing a couple of books. Again, don’t believe me. Believe The New York Times, The Washington Post, USA Today, every investigative reporter in the United States. Bilingual education does not work or has never worked anywhere in the United States or in the world on a large scale.
CATHERINE SNOW: First of all, let me say in California it might be the case. I try not to talk about things I don’t know a lot about [applause], that only Latino children are receiving bilingual education. Right here in Cambridge, I would point out, there is a maintenance bilingual program in Chinese, one in Korean, one in Creole, one in Portuguese. There is a two-way program in Spanish. There are Chinese and Khmer and Japanese and Russian and Hebrew bilingual programs within 10 miles of where we are right now. So that is simply not true in Massachusetts. And it’s one of the reasons why making educational policy by referendum is really a bad idea. You get a one-size-fits-all analysis of the problem and a one-size-fits-all purported solution to the problem.
So let’s talk about the test scores in California. Now, I notice you’re talking about the test scores. Why aren’t you talking about the reclassification rates? The reclassification rates which-remember the claim, 95 percent failure because only five percent reclassification? Well, since 1998 and the introduction of intensive immersion programs for children in California, the reclassification rates have soared to seven percent. Right now the average reclassification rates across the state of Massachusetts range from 17 to 25 percent in different school districts. What that means is that in Massachusetts, on average children are getting out of bilingual programs in four to five years. In California evidently they were not. Once again Massachusetts and California are different places. Perhaps we do not need California’s initiative to solve Massachusetts’ educational problem.
Now about the test scores. It’s true, a million LEP children in California have seen very significant increases in their test scores over the last three years. So have several million non-LEP students in California. The fact of the matter is that five years ago California ranked 48th in the nation in education achievement. This is a position from which it is not hard to raise your test scores. The Governor of California at that point decided he would invest a lot more money in education. Statewide standards were put in place for literacy and math teaching. New tests were introduced; the SAT-9 was introduced in 1998. Class size was reduced in primary school. And Proposition 227 was passed.
Which of these many changes is responsible for improvement in educational outcome? Well, if it was only LEP children in California whose test scores had gone up we might say it was Proposition 227. The fact of the matter is that everybody’s test scores in California are going up. In fact if you look across the state of California at kids in intensive immersion and kids who are still in bilingual programs, and native English-speaking children or children who are not requiring special attention because of language learning, you will find that all of their test scores are going up more or less equivalently. The gap between the native English speakers and the non-native speakers is not decreasing, unfortunately, but the difference between intensive immersion and bilingual education is actually in favor of bilingual ed programs.
Now, Mr. Unz, who seems to take his data from newspapers rather than the state scores, talks about Oceanside. Well, for every Oceanside, I have to tell you, there’s an Oceanview. Oceanview is a district which also moved from bilingual education to intensive immersion programs. And there the test scores haven’t risen at all.
The fact is that the SAT-9, which is not a test that’s designed for second-language speakers of English, is a test on which you can double your score by getting one or two more items correct when you’re a five-year-old. It is a very poor test. It does not reflect the kinds of things we really want kindergartners and first-graders to be learning. They need to be learning lots of language, and they need to be learning the skills that will enable them to read, to reason mathematically, to do science. Those are not skills that the SAT-9 tests. And even if the scores really are higher, it doesn’t matter at all; it does not reflect educational outcomes.
The outcomes that we want to look at are learning and achievement, comprehension, reasoning, math, and science. Let’s look at reclassification rates. I’m in favor of reclassification. I think kids should be in mainstream programs within three or four or five years. That’s something that Proposition 227 in California has not impacted at all.
TONI RANDOLPH: I’m just going to ask you to field a couple of questions before we turn it over to audience questions. Proposition 227 has only been in effect in California for three years. That’s not a very long time. Even though you have cited some successes, it’s not a long time. How do you know whether over the long haul, whether it’s three more years or 30 more years, it will continue to have the same success it has had?
RON UNZ: You have raised a very good point. During the campaign for 227, which occurred in 1998, the opponents of the initiative again refused to defend the existing status quo. They said, “Oh, we know we have problems in the schools.” But they said if the measure passed the result would be a catastrophe. Test scores would plummet. These very low test scores would get much worse. Instead-I’ll repeat this one more time-the average test scores of over a million immigrant students have gone up by 40 percent in less than two years and continue that way. So it’s now 50 percent in less than three years.
Furthermore, those school districts, when you look at the state of California, that tried to keep their bilingual programs, showed little if any improvement. Those that got rid of their bilingual programs showed dramatic gains. The 40- or 50-percent average includes a number of 100-percent gains and a number of 10- or 15-percent gains in those districts that resisted the initiative.
It didn’t only persuade people who already believe in the issue; it persuaded many of the people who opposed that issue. Now, if we’re talking about 10 or 15 or 29 years down the road, it could be that if you take young children and teach them English when they’re in kindergarten, when they reach the age of 40 or 50 their brains explode and we’ll have terrible catastrophes happening in 2020 or 2030 because of this initiative that we implemented now. But when we’re doing any sort of scientific analysis, we have to look at the reality that is happening. The reality is that the first year test scores went up by 20 percent, the second year they went up another 20 percent, and now in the third year they have gone up by about another 15 percent.
We are talking about no evidence whatsoever of any adverse consequences. The parents are very pleased, the children are doing very well in school. They’re well adjusted. They seem very happy. They are learning English, and they are learning exactly the same sort of English that all of these other immigrant groups were learning.
Now, it is true Massachusetts is one of the very few states that has bilingual programs for non-Hispanic students. Virtually everywhere else in the country, bilingual programs are almost solely for Hispanic students. And all the non-Hispanic immigrant students have always been taught English during this 30-year period and have done very well in it. Now, maybe there’s some mysterious thing that means Hispanic students need bilingual education none of the others do. But there seems no evidence for that whatsoever. The point is, if you get rid of these programs, I think that Hispanic students will do just as well as all the non-Hispanic students.
TONI RANDOLPH: California does have one of the largest Spanish-speaking populations in the country. And the same is true in Massachusetts. How do you compare programs? How much [of a] study do you need? How much time do you take to ensure that a program is successful? How do you compare [a program] to the program that was in place? Is there a way to do that, to say this is successful and this was not?
CATHERINE SNOW: Ideally, of course, you would do a proper experiment. You would randomly assign children to structured immersion or a bilingual program. And, you would make sure that the structured immersion program was as good as it could possible be and that the bilingual program was as good as it could possibly be. We can’t do that kind of random assignment study. So, in fact, what ends up happening is that people make inferences from comparing fourth graders now and fourth graders five years ago. You’re not looking at the same children; you’re looking at different cohorts of children.
Another very good design would be to actually look at children, follow them through their schooling. And that’s what I’m proposing will ultimately reveal the negative consequences of structured immersion for the children who are in it now and were in it last year. Of course at the end of the year [a student] might know some more words in English, but [has that student] acquired sufficient English competence to actually deal with the academic challenges of third and fourth and fifth grade school? You have to do the longitudinal study in order to be able to see that effect.
Now, once again, let me say that Mr. Unz keeps saying a million LEP children have increased their test scores enormously in California, and that’s true. So have the non-LEP children; so have the LEP children in bilingual education. The comparison that he is offering you is a comparison based on a couple of poster school districts and not looking across the entire state of California at an honest comparison of all the children in English immersion and all the children in bilingual education.
RON UNZ: If I can respond to that, Professor Snow is simply in error on that point. There have been studies done across the state of California examining all of the million students we’re talking about. In fact, probably the best study was done by the San Jose Mercury News, a very well regarded paper that had opposed the initiative. It had opposed Proposition 227, urged people to vote against it. After the initiative passed, it spent several months after the first year of test scores doing a statewide quantitative analysis of all the students in the state of California, those who stayed in schools that kept their bilingual programs, those who shifted to intensive English immersion programs with rapid mainstreaming. And the result they proclaimed, in a banner headline on the front page, was that the students in the English immersion schools and districts showed test scores depending on grade and subject level that were 20 percent, 40 percent, even 100 percent higher across a million students statewide than the ones who stayed in the bilingual program, and that was after one year.
Or take, for example, two neighboring school districts, a perfect controlled experiment. Oceanside and Vista are both medium-sized school districts down near San Diego. They both have heavily Latino immigrant populations, very similar demographics, similar size, everything about them is the same. The advocates of bilingual education always pointed to Vista at the time of the campaign and afterwards as having some of the best bilingual programs in the state, [schools] that were keeping their bilingual programs despite the initiative.
Oceanside was always denounced by those people as the district that most completely got rid of all their bilingual programs and was doing everything wrong. After two years of test scores, the Vista scores had moved up virtually not at all. There had been virtually no improvement among immigrant students in Vista in reading, while the Oceanside test scores had doubled.
Now you have two neighboring school districts, very similar; both groups on both sides admit that one of them is bilingual and one of them is anti-bilingual. One of them shows very little improvement at all and the other doubles their test scores. That’s a very useful piece of evidence and, again, a statewide analysis of all the million students showed dramatic gains for the non-bilingual class students as opposed to the ones that stayed in the bilingual district. The facts are in. It’s absolutely clear-cut. And that’s the reason all of these people I was talking about in California have switched their sides on the issue.
TONI RANDOLPH: What does the evidence tell you?
CATHERINE SNOW: I’d like to ask Mr. Unz a question. When the San Jose Mercury News reported cold fusion, did you believe it? Do you think that is the test for data analysis and drawing conclusions? There are statisticians. There are psychometricians. There are researchers in California who give a totally different reading. I’m supposed to believe the San Jose Mercury News?
RON UNZ: Okay. On that point, I hate to get personal here, but . . . We are talking about a program, bilingual education, that has been imposed in the United States and supported for decades by the educational theoretical establishment. Endless numbers of professors of education believe in bilingual education, but I would argue that the number of academics, professors of education, who support bilingual education outnumbers those who oppose bilingual education by probably about a thousand to one. If you go to all the professors of bilingual education and ask them whether bilingual education works, they will say, “Of course it works. I’m a professor of bilingual education. I believe in bilingual education. I’ve always believed in bilingual education."
Unfortunately, the reality is contrary. The analysis the Mercury News did was not a complex analysis. If you simply look at the test scores of those students in districts that kept their bilingual programs statewide and you compare it to the test scores of those students in districts that got rid of their bilingual programs, you see a gigantic difference across the state of California in a million students. Reality trumps theory. And to be honest, in past decades, in past eras, there were many professors of theoretical physics who were totally wrong. They believed in nonsense, and after a while they all died off, and a new generation of professors arose that believed in reality.
What we are seeing now is the collapse of the ideas of a gigantic group of theorists who believed in bilingual education for 30 years. And that belief and advocacy destroyed the lives of millions upon millions of students in America during that period. That is what the history books will write.
CATHERINE SNOW: Hey, Ron, we’re all going to die. You too.
RON UNZ: Have we crippled the students who have come up in the last 30 years?
CATHERINE SNOW: Of course not. Once again, I am not defending poor educational programs. I cannot defend poor bilingual programs, and I certainly cannot defend poor immersion programs. They are out there on both sides. Program quality trumps program type, I’m quite sure that’s true. So of course if a school district determines that it’s going to put a lot of energy into improving its reading instruction, improving its math instruction, and teaching its children English, it can get improved outcomes if it continues to do that with a structured immersion program, or it can choose to do that with a bilingual program.
Program quality is very important, but I could mention dozens of students I know, for example, who have gotten into Harvard who are products of bilingual programs. These are people whose lives have been enhanced by bilingual programs. I could take you down the street and show you two-way bilingual programs where not only Spanish-speaking children but English-speaking children feel their lives are being enhanced by the opportunity to develop bilingual skills from kindergarten on through elementary school. The notion that good bilingual education is bad for children is simply nonsense.
TONI RANDOLPH: Is it a district’s commitment to bilingual-and bilingual can mean a lot of things, whether it’s immersion or the traditional bilingual-that makes the biggest difference?
RON UNZ: I would argue again very simply and straightforwardly that bilingual education has never worked anywhere in America on a large scale in 30 years. With any sort of theory, you can always find a few isolated cases where it might work under ideal laboratory conditions. You can find a classroom where it works. You might be able to find a school where it works, allegedly. You might be able to find a very small school district where it works, but there has been no large-scale example anywhere in America in 30 years where bilingual programs worked. And I’ve repeatedly asked the advocates of bilingual education to point to any large school district anywhere in America that they can cite as a successful example implementing their programs, and they can’t think of one.
Now if something has never worked anywhere on a large scale in 30 years, maybe it just doesn’t work. And it is true that you can find dozens of students as Professor Snow cited who benefited tremendously from these bilingual programs. I believe that, but I think you can find millions of students who are hurt by these bilingual programs, and you have to go with the numbers.
In the state of California, again, those school districts that were cited by bilingual advocates as having some of the best bilingual programs-like Vista, that were doing everything right-they showed virtually no gain in test scores in the period following the initiative. By contrast, those districts that got rid of their bilingual programs showed enormous improvement, and let’s look at other examples.
Everybody knows that a lot of Latino immigrant students do a lot better in the Catholic schools than they do in the public schools. Many of them, in fact, if they were given the opportunity, would probably rather send their children to Catholic schools, parochial schools, rather than the public schools. Catholic schools do not use bilingual education. The students are taught English right away. Now if these same immigrant students do so much better in Catholic schools than they do in public schools without bilingual programs, couldn’t part of the reason be that they’re being taught English rather than being taught Spanish?
And, again, when we’re talking about theories: Anybody here in this audience who believes that the older you are, the easier it is to learn another language should vote against our initiative, because that is the basis of these bilingual programs, which is utter nonsense.
. . . How long do you think it takes a young child to learn English at five years old?
CATHERINE SNOW: I think to learn the first words could take half an hour. I think that if a child in an immersion classroom is learning five words of English a day, that’s probably pretty good. I think by the end of a year the child could know a thousand words of English, could have a wonderful conversation with an avuncular adult who comes in and says things like, “What’s your name and where do you live and what kind of games do you play?” I think, however, that at the end of one year that child will not be sufficiently proficient in English to have an easy time learning how to read in English. There is a very high risk of trying to teach children to read in languages they do not speak well, and reading is the ultimate test. If children are not learning to read well, school is not doing its best by them.
RON UNZ: That’s an excellent statement of the case for bilingual education and the theory behind it. The theory behind it is that it’s wrong to teach children English until you spend sometimes years teaching them in Spanish, including reading and writing in Spanish.
TONI RANDOLPH: I think our audience really wants to chime in here. I’ll start on this end. I’ll defer to our friend, [Boston University chancellor] John Silber.
JOHN SILBER: As a matter of fact, there are a number of theories propounded with regard to bilingual education. In neurophysiology, we have found that the young child up to the age of about seven is learning the foreign language in exactly the same area of the brain in which the native language is acquired, and that means it’s a process by which no translation is necessary. You can see this example in Switzerland. You can see this example in the Netherlands, where children are exposed to maybe three or four languages, and the results are there.
You also have theories propounded by professors that you can’t learn a second language until you’ve mastered the native language. If that were true, no child would ever speak, because they have to learn the native language before they ever learned any language. That’s the kind of nonsense that has been put out by some professors of education who simply don’t know what they’re talking about.
What I would like to know is how can any responsible people be so indifferent to the facts. When you find a differential result in the presence or the absence of bilingual education and it favors those that have learned in the absence of bilingual education, how can anyone who is a responsible intellectual simply ignore the results of a successful examination using those methods of difference?
CATHERINE SNOW: If I believed that the data in California showed that programs that had adopted intensive immersion were now producing better results than programs that had bilingual teaching methods, I would of course concede that Mr. Unz had proven his point. I do not believe that. The data do not demonstrate that. The findings do not support that conclusion.
AUDIENCE MEMBER: [Says her child has benefited from a bilingual education program.]
RON UNZ: Good. I’m glad your child is doing so well. That doesn’t mean that your child might not have done even better if your child hadn’t been taught English immediately. Now, again, when you’re looking at it, you have to look not at one example or another example, one anecdote or another anecdote, but the result across a country or the result across a state. And the evidence when you’re talking about thousands or tens of thousands or hundreds of thousands of students is that the students who are given bilingual education do much worse than the students who are not given bilingual education, and that really is the way general policy has to be made.
With regard to your child, our initiative in Massachusetts that will be on the ballot in about a year would allow parents like you, who really believe in bilingual education, to apply for a waiver, to place or keep your children in that program. So, in other words, maybe your child is a special child who actually would benefit from bilingual education rather than be hurt by it. If so, then you can apply for a waiver and keep your child in that program.
AUDIENCE MEMBER: If we don’t want this to happen, how can we prevent it from going through?
CATHERINE SNOW: I would say that if you don’t want it to happen or if you do, you go to the poll and vote yes or no.
RON UNZ: With regard to your question, there is a very interesting dynamic which I learned about this issue during the California campaign. All the public opinion polls in California over a year showed that it was one of the most popular initiatives in the history of California. Yet when I attended public forums or debates like this one, usually the audience was 95 percent against it, sometimes it was 98 percent, sometimes 100 percent. And that’s simply because the people who are most motivated to attend public forums dealing with bilingual education tend to be advocates of bilingual education, including a lot of bilingual teachers and bilingual administrators and bilingual academics.
The reason this program has stayed for 30 years even though it’s never worked and has been incredibly unpopular is because it’s supported by a very small, but very vocal and committed, group of people and opposed by the overwhelming majority of the rest of society who is home watching television. So in other words, the people who came to this forum right now believe in bilingual education. I would assume that’s most of you, based on your headbands. The people who will vote against bilingual education in a year’s time are home watching television right now. It’s not a matter of money.
For example, in California we were outspent 25 to 1 in advertising. The supporters of bilingual education spent 25 times more money in advertising than we did, but we still won in a landslide. The same thing [was true] with Arizona, and the polling numbers in Massachusetts show that support for getting rid of bilingual education is far stronger here than it ever was in California.
CATHERINE SNOW: Can I just answer this question briefly? One of the reasons of course is that Ron Unz is a very good communicator. And because he presented this issue in terms that misrepresented the facts-he called the campaign, “English for the children”-he suggested that bilingual education is not in favor of providing education for children. If you ask any immigrant parent, “Do you want English for your child?” Of course the answer is going to be yes. That’s part of the reason why this conversation is very important. It’s necessary to make clear to the public what this issue really is.
RON UNZ: If I can respond, it sounded like what you were saying is that the Asian students in California who have never had bilingual education were doing very badly in school. They are actually over-represented at California universities by a huge fraction. So, for example, Asian students make up about 10 percent of California’s population, but are almost 50 percent of the students at the top universities. So even though maybe some Asian students aren’t doing well, most of them are doing very well. In fact, they’re doing as well or better than the Anglo students.
I’m not saying the only reason is because they never had bilingual education. But the fact that they’re doing so well in universities and they never had bilingual education reinforces my case that maybe one reason that Hispanic students in California are doing so badly is that they had bilingual education.
AUDIENCE MEMBER: I was interested in what you [Mr. Unz] said about theory and reality, because the theory that you had is that this test kind of addressed people equally across the board without considering their cultural backgrounds. I’m wondering what you think about the tests and how they address these different backgrounds?
RON UNZ: You’ve raised a very good point. Obviously the Latino students, for example, in California tend to be disproportionately over-represented as being from poor families or poorly educated families. For example, in Oceanside and in Vista, the immigrant students in both those districts tend to come from poor farm worker families. So they certainly have a lot of cultural and educational disadvantages.
The tests basically involve reading, mathematics, subjects like that. I think most of the parents are much happier when the students double their test scores on the tests than if they don’t do well. And that’s why they’re happy about the initiative.
AUDIENCE MEMBER: I wanted to ask about older students.
RON UNZ: You’ve raised a very good point. Our initiative makes a clear distinction between older students and younger students. In other words, students who are 10 years and older would have a much easier time getting a waiver to be placed or kept in a bilingual program as opposed to English immersion.
Now, again, the theorists who believe in bilingual education actually think that the older you are, the easier it is to learn another language. So they would say older students would have an easier time in English immersion than younger students, but I think that’s nonsense.
The point about it is that under our initiative, we make that distinction. And, you know, I’d like to say I actually have talked with a lot of immigrants. I know a lot of immigrants. I work with them. I know a lot of them. I’ve actually asked them their opinion about English immersion even for older students, because many of them arrived in the United States when they were 12 or 13. Most of them seem to believe that even at that age, it’s better to be intensively taught English than to be taught academic subjects in your native language while you’re learning English.
So there’s a division of opinion on that, but I do believe at least that you can make a case in favor of bilingual education for older students. While I think there’s no case in the world, no logical case in the world, as to why you don’t teach a five-year-old immigrant child English immediately once they start kindergarten, and that is the key issue involved.
There’s one very important statistic. Over half of all the immigrant students in the United States who are classified as not knowing English were actually born in America. Most of the remainder came here when they were very young, so the vast, vast majority of so-called limited English students actually started school in the United States when they were five or six years old. There were relatively few who came here at 12 or 13. The reason that they don’t know English when they’re 12 or 13 is many times that they’ve been in these Spanish almost-only classes for six or seven years, and that’s the problem.
CATHERINE SNOW: I’m sorry. I’m going to have to admit here that I’ve actually collected data on the speed of language acquisition among older and younger second-language learners. I mean testing the kids, going to see them every six weeks, and figuring out how much they learned in these six-week intervals.
My data are pretty clear: 12-year-olds are much faster language learners than five-year-olds. So although you might ridicule this perspective, I can’t abandon it because it’s not a theory; it’s actually based on hard data. So my sense is that older learners can in fact acquire a second language quickly enough in an immersion setting. And I think if you’re giving waivers, easily you should give the waivers for the children under 10, not over 10.
AUDIENCE MEMBER: I’m not a bilingual educator. I’m not an educator. I’m a journalist who [writes about] social policies. I want to take this from a social policy angle and talk about politics. My definition of politics is two questions and four words. Who benefits? Who decides? Now, most of the social policy, education policy in this country in the past 30 years has been based on the assumption of a deficit model of urban education. And, unfortunately, funding tended to match that deficit structure.
Taking the data that you put out in terms of the failure rate for bilingual ed, the question for me is: What factors are relevant to look at that? Now, there are a number of factors, including funding programs. But I want to kind of jump really quickly to an analogy that I think applies. Suppose I am trying to prove that nuclear fission works. Try to get fission by striking two pieces of pitchblende together. I don’t think you’d find fission. And I suspect that the way that you’re interpreting data is pretty much the same; you’re striking two programs together hoping for a spark and there’s not a spark.
RON UNZ: Well, to the extent that you were talking about who benefits from these programs, that raises a very important issue which I should respond to. We have what might be called “the bilingual education industry” in the United States. In other words, you have the bilingual teachers and the bilingual administrators, the bilingual academics, the bilingual textbook manufacturers, the bilingual coordinators. They benefit from bilingual education whether it works or it doesn’t work. They are the mobilized groups that keep this program in place even though it doesn’t work.
Another example of somebody who benefits-we were outspent again twenty-five to one on advertising in California. The