Whole language instruction has been an educational disaster in California. It is entrenched, however, because it was the only method of reading instruction presented in the universities and colleges that had teacher certification programs.
Quotes from an exellent two-part article written by Jill Stewart, published in the LA Weekly in 1996.
In 1987, whole language theory began its sweep across California in the form of a nationally acclaimed reading "framework" adopted by the state Board of Public Instruction that downplays the teaching of traditional reading skills. "The core idea of whole language," says one of its most vocal proponents,
Mel Grubb of the California Literature Project, "is that children no longer are forced to learn skills that are disembodied from the experience of reading a story. The enjoyment and the wonder of the story is absorbed just as the skills are absorbed."
The central tenets of the philosophy hold that small children trained with such techniques will write more expressively, love reading, fully consider whole meaning over mere words, and emerge as more sophisticated readers, writers and thinkers.
But whole language, which sounds so promising when described by its proponents, has proved to be a near-disaster when applied to--and by--real people. In the eight years since whole language first appeared in the state's gradeschools, California's fourth-grade reading scores have plummeted to near the bottom nationally, according to the National Assessment of Educational Progress(NAEP). Indeed, California's fourth graders are now such poor readers that only the children in Louisiana and Guam--both hampered by pitifully backward education systems--get worse reading scores.
In the second part of the article, Stewart combines moving vignettes with a succinct outline of how California's educational structure was highjacked by Whole Language true believers.
[snip] or fell behind because the teacher was preventing them from sounding out their letters were labelled by reading specialists as "slow readers" or "learning disabled."
Teachers related tales to Joseph in which, "if they tried to teach phonics or word attack skills to the kids who weren't getting it from the storybook and the invented writings, bureaucrats came in from their district office and ordered a stop to it. It was terrible stuff, virtually a new religion, a cult."
In the end, a rudderless group of state officials were left struggling to interpret a unique and untested reading philosophy which they, themselves, did not understand. At the schools, deep divisions broke out as district bureaucrats began dictating bizarre orders to teachers and principals.
A meeting of top state curriculum officials was called in 1993. There, whole language "true believers"--including the powerful California Reading Association, California Literature Project and several state officials--successfully deflected an attempt to re-emphasize basic skills in gradeschools. According to those who attended, state education officials Dennis Parker and Fred Tempest argued that teachers would "go nuts" if asked to make another big change in reading methodology. Teachers, they insisted, merely needed time to absorb whole language's unusual techniques.
Is it any better now? Not really. Bill Carlson wrote an update in 2001: Another Status of Education: Still Bungled.
What’s the status of WL now during June 2001: Well, most of the WL devotees are still in key positions and still making the big decisions. I continue to believe that top dishonors for California’s literacy crisis go to the California State University System for producing so many improperly trained reading teachers. Often ignored mid 1990’s legislation that are still “on the books” mandates direct, systematic, and explicit reading instruction. Such tried and tested instruction continues to work well for children in private schools, home schools, Sylvan Learning Centers, and public schools with a serious focus on literacy; all seem to do quite well without WL enhancements.
In 1995, there was a movement to adopt Whole Language in the Massachussets public schools, which encountered resistance.
A group of 40 linguists, psychologists and other academics who claim that whole language has been a disaster where it has been tried elsewhere -- particularly in California, where reading scores went down after a whole-language curriculum was adopted.
"As linguists, we were in a position to see that rotten science is at the heart of a system being used to educate countless children across the country, including our own," said David Pesetsky, a professor at the Massachusetts Institute of Technology who has children in the Lexington schools. Pesetsky also signed the "group of 40" letter of protest to Education Commissioner Robert V. Antonucci in July 
In 2000, the News-Gazette ran a roundup of educational trends
Reading education: The best way to teach children to read has been a major issue across the country. The two methods in question: whole language vs. phonics. Tired of the endless drills associated with phonics, many schools went to the whole language approach. Proponents touted it as the best way to teach kids to read, but others called whole language an "educational disaster" and blamed it for pulling down reading scores. The reading wars died down a bit after the influential National Research Council issued a report in 1998 urging more teaching of the connection between letter and sound, or phonics, as well as exposing students to a wide variety of reading material. Whole-language advocates believe kids are turned on to reading by exposure to lots of books, allowing them to get meaning from context.
The WorldNetDaily is a conservative site; the article quoted below is by Samuel Blumenfeld. It will start to give you a feel about how whole language vs. phonics has become politicized. If you believe in whole language, you are likely to be on the liberal-to-socialist spectrum; if you believe in direct phonics instruction, you have to march in the same parade as Phyllis Schafly, the Eagle Forum, and Dr. Blumenfeld.
The simple truth is that phonics has been politicized by the left ever since it became identified with conservative educational principles and practices. But this is by no means a recent development. It really all started back in 1898 when John Dewey wrote his famous essay, "The Primary-Education Fetich," in which he advocated shifting the emphasis in primary education away from the development of academic skills, particularly reading, to the development of the social skills. [snip]
Phonics teaches a child to read what the author wrote, not what he thinks the author wrote. Today's anti-phonics, whole-language teachers basically [deny that the author creates meaning]:
Whole language represents a major shift in thinking about the reading process. Rather than viewing reading as "getting the words," whole language educators view reading as essentially a process of creating meanings. Meaning is created through a transaction with whole, meaningful texts. It is a transaction, not an extraction of the meaning from the print, in the sense that the reader-created meanings are a fusion of what the reader brings and what the text offers. ... In a transactional model, words do not have static meanings. Rather, they have meaning potentials and the capacity to communicate multiple meanings.
No wonder children are having such a tough time learning to read in American schools, and no wonder parents want to get back to phonics.
The following is from LD Online, a site I highly recommend. The reviewer is responding to a book by Dianne McGuinness, who is touting her own proprietary method of teaching the decoding piece of reading.
McGuinness's criticisms of current methods of reading instruction---both whole language and traditional phonics---are generally on the mark. Whole-language programs that encourage children to guess at words based on context and that shun decoding instruction are a disaster for many children, especially for poor readers; traditional phonics programs often teach unnecessary terminology or rules, use confusing language, and may take forever to get through.
However, I do have three substantial concerns about the book. First, especially when discussing the details of her reading program, McGuinness's tone is less that of a scientist than of a political revolutionary. This sort of dogmatic rhetoric is already rampant in the field of reading education and only tends to perpetuate the endless pendulum swings that have long characterized the field.
[snip] [Second,] McGuinness appears to overemphasize decoding at the expense of reading in context and comprehension. Most beginning readers, and many older poor readers, do benefit from an emphasis on decoding, including some work on decoding words in isolation.
Finally, although I could not agree more with most of McGuinness's criticisms of the concept of "dyslexia," I think she gives individual differences in the abilities that underlie reading (e.g., phonological abilities) too short shrift.
[T]here are substantial individual differences in the ease with which children learn to decode---even with good instruction. Some children need much more intensive, lengthy instruction in order to learn to decode words than do others, and scientists are just beginning to understand the type of reading instruction that might be optimal for these children. As we struggle with much-needed changes in educational policy, it is very important that we maintain a variety of instructional options to accommodate these individual differences in children.