The most difficult task facing state policymakers is to understand that the strong feelings on both sides of the reading debate may not be truly reflective of the choices before them. It is not a choice between teaching reading through phonics or whole language. A good whole-language program must include phonics. And good direct-phonics instruction, while essential, is only one part of an effective reading program.

Between 1992 and 1994, the performance of North Carolina fourth-graders on the National Assessment of Educational Progress (NAEP) reading test showed measurable improvement. The 1994 results were slightly above the national average and tops among the 15 member states of the Southern Regional Education Board.

While the NAEP results showed that some progress had been made, they also suggested that North Carolina schools, like those in most other states, were failing to teach many children to read at acceptable levels. Despite the gains over 1992, the 1994 NAEP results found that 41 percent of the state's fourth-graders
still were reading below the basic fourth-grade competency level. In the spring of 1995, legislation was introduced in the North Carolina General Assembly that was intended to make dramatic changes in the way reading was taught in the state's public schools.

The situation in North Carolina was representative of what was happening across the nation. Parents and policymakers saw unacceptably large numbers of children completing elementary school and even high school without having learned to read well. The initial response in many states was a movement to throw out the predominant approach to teaching reading, generally referred to as whole
language, and replace it with something completely different. In most cases, that something completely different was referred to simply as phonics.

• Teachers have the rug yanked out from under them.

The debate that pitted phonics against whole language seemed discouragingly familiar to many longtime observers of public education. Historically, approaches to teaching not just reading but virtually every other aspect of education in kindergarten through grade 12 have tended to follow the pattern of a pendulum that swings periodically from one extreme to another and rarely comes to rest anywhere near the middle. Teachers are resigned to the fact that they can expect to do things one way for a while and then, just as they are beginning to get the hang of it (and often before there has been time to evaluate properly the impact of the latest fad), the rug will be yanked out from under them and they will be directed to do something completely different.

The sudden swing of public opinion toward phonics and away from whole language was typical in that it was characterized by widespread misunderstanding about what the two ideas, whole language and phonics, were about. It was also typical in that it reflected an unfortunate tendency, whenever some new approach turned out to be less than perfect, to want to abandon it and go back to something that supposedly had worked much better "in the old days."

Phonics and Whole Language

The term "phonics" refers to a method of reading instruction that focuses on instruction in the "building blocks" of written language - knowledge of the alphabet and of the sounds (called phonemes) formed by letters individually and in combination. Few on either side of the debate would argue that mastery of these building blocks is not essential to learning to read. The disagreement arises from the question of how they should be taught.

During the 1970s, the most prevalent model for teaching reading was what sometimes is referred to as "drill-and-kill" phonics, which emphasizes repetitive, rote learning exercises and worksheets, with actual reading for meaning taking a secondary role, especially in the critical early elementary years.

While phonics refers to specific skills that are an essential part - though only part - of learning to read, whole language represents a broad philosophical approach to the teaching of reading. Building on new knowledge and understanding of how children develop and learn, it assumes instruction should be responsive to the child's individual needs and should draw upon the child's own interests and desires as motivation for wanting to learn.

Fundamentally, whole language believes that children need to be exposed to large quantities of quality literature and that all aspects of the curriculum, whether math or science or social studies, should be viewed as opportunities to teach reading skills. Most whole-language advocates would agree that the skills associated with phonics should be taught as part of a comprehensive reading program. In contrast to the drill-and-kill model of phonics instruction, however, whole language tends to embed phonics in reading for meaning, teaching skills only as the opportunity arises or as children demonstrate the need for a particular skill.

• Neither instructional model can meet the needs of all children.

In fact, it has become clear that neither drill-and-kill phonics nor a pure whole-language program can meet the needs of all children. Ironically, the rise of whole language as the preferred approach to teaching reading during the 1980s was fueled by the dominance and the perceived misuse of phonics during the 1970s. The emergence of whole language as an alternative instructional model was a response to the fact that, while many children did learn to read with drill-and-kill phonics instruction, others did not.

By the late 1980s, whole language had displaced phonics as the dominant model for reading instruction in many schools. Unfor-tunately, while many children prospered under whole language instruction, many still failed to become good readers. By the mid-1990s, the failures of whole language had left many people longing for the "good old days" when phonics was king, forgetting that the drill-and- kill phonics approach had been rejected precisely because it too had failed many children.

For support, advocates of a phonics-based approach cited new scientific research showing that awareness of letter-sound relationships is the main tool used by good readers for decoding unfamiliar words. The research clearly showed that a majority of children who fail to become good readers do so because they have not mastered these letter-sound relationships.

• Amount of phonics that students require varies.

Unfortunately, many phonics supporters neglected to mention that the research also showed that direct instruction in letter-sound relationships is most effective if it is done as part of a comprehensive literature-based reading program having many of the characteristics of whole language. The research also made clear that the amount of direct phonics instruction different children require varies and should be tailored to individual needs as much as possible.

Many whole-language advocates, on the other hand, refused even to consider the implications of this new research, arguing that the experience of the 1970s had discredited all types of direct phonics instruction once and for all.

What those at the poles of opinion in both camps have failed to recognize is that while on the one hand the research shows that phonics is an essential part of any effective reading program (and that some children will master the necessary skills only through direct instruction), on the other it also confirms that phonics is only a part of a good reading program. A reading program that holds out phonics as the only way to teach reading is no more a complete reading program than a whole language approach that fails to address phonics.

North Carolina's Struggle to Find Balance

During the 1980s, a number of states (most notably California) made a strong commitment to using whole language in all public schools. In contrast, North Carolina public schools have tended more than most states to cover the full range of the spectrum in terms of the instructional models they use. By the late 1980s, while some school systems in the state had committed to using whole language, many others still were using something closer to traditional phonics.

In 1988, the legislature, responding to significant advances in knowledge of how young children learn, passed legislation that prohibited schools from using state funds for standardized testing in kindergarten through grade two. Instead, the state Board of Education and the Department of Public Instruction were directed to develop new assessment models for reading and math that could better provide meaningful information on which to base decisions about children's individual instructional needs.

The new reading assessment reflected the department's belief that many schools were overemphasizing phonics at the expense of reading for meaning. One of the primary ob-jectives of the assessment was to move schools toward a more individualized, literature-rich model of reading instruction.

• No state funding was provided for teacher training.

The concepts on which the new assessment was based were unfamiliar to many teachers, so the department embarked on a staff development program to help school personnel understand how to use it. Unfortunately, no state funding was provided specifically for this purpose. (School systems in North Carolina generally receive state staff-development funds directly, to use as they see fit.) Initially, the department diverted funds that previously had been used for standardized tests - approximately $100,000 a year for two years for both reading and math - to pay for the staff development effort. When these funds were eliminated by the legislature, the department was forced to charge registration fees for training.

The situation was complicated by the fact that the 1988 legislation, while banning the use of state funds for standardized tests in K-2, did not require all school systems to use the new assessment model. This fact, combined with local school systems' discretion in using their staff development funds, meant that implementation of the new assessment was extremely uneven across the state.

• It was easy to lose sight of phonics.

Further complicating the picture was the fact that the assessment only implicitly reflected the importance of word recognition skills -including awareness of letter-sound relationships. This meant that for those schools that had moved toward the whole language end of the spectrum it was easy to lose sight of the idea that a good whole-language program always should include attention to mastering essential skills.

Several bills introduced in the North Carolina General Assembly in 1995 reflected a belief that the state's schools were dominated by whole language, when in fact there was relatively little consistency either in methods or in quality of instruction from system to system. The proposed legislation essentially would have mandated that phonics would be the only acceptable approach to teaching reading in the state's public schools and colleges of education.

No reading legislation won approval during the 1995 session, but the Department of Public Instruction responded to the concerns raised in the legislature in 1995 by revising the K-2 assessment to include more specific references to the importance of word recognition skills. The changes were insufficient, however, to satisfy critics who had come to view the assessment as essentially a document that promoted whole language.

Staking Out the Middle Ground

In its 1996 session, the legislature revisited the debate over how to respond to the problem of reading instruction. Legislation carried over from the 1995 session to make phonics the only acceptable way to teach reading in North Carolina schools was again debated. Substitute legislation was finally enacted that called for "the implementation of balanced, integrated and effective programs of reading instruction."

The new legislation directed the state Board of Education to develop "a comprehensive plan to improve reading achievement in the public schools ... based on reading instructional practices for which there is strong evidence of effectiveness ..." The legislation went on to make clear that the plan should "include early and systematic phonics instruction." The Board of Education was given six months to develop its plan and report to the Joint Legislative Oversight Committee.

• Plan clearly defines balance.

In January 1997, the state Board of Education and the Department of Public Instruction presented a draft plan to the legislature that addressed concerns on both sides of the phonics/whole language argument. Following further revisions in response to comments from legislators and the public, the plan was adopted formally by the board.

The final plan clearly states what it means by balance: "Efficient early instruction contains a balance of activities and strategies to improve word recognition, including phonics instruction, reading meaningful text, writing and spelling activities. Effective teachers interweave these activities in their instruction and make sure that direct teaching of skills is done in conjunction with reading connected, informative, engaging text."

Consistent with the plan, the North Carolina Standard Course of Study for English Language Arts has been revised to provide detailed guidance on how to teach phonemic awareness and other word-recognition skills within the context of a literature-based reading program. At the same time, the grade-level benchmarks developed by the department to help in assessing children's progress in reading and writing have been revised again to reflect the curriculum changes.

Teacher Training Essential to Success

The North Carolina plan also addresses the need for both preservice and inservice professional development if the proposed curriculum changes are to be implemented effectively.

"In teacher education programs, additional competencies specifically addressing the role of phonics in reading are needed. Field or clinical experiences in reading are needed for elementary teachers. Existing elementary teachers need additional training in incorporating phonics into a balanced reading approach."

The revisions to the standard course of study were developed with the active involvement of university reading specialists, and those revisions form the basis for a review of reading instruction in all college and university programs preparing elementary, middle, secondary and special education teachers. During the 1997-98 academic year, each campus is to approve curriculum revisions in its teacher education program designed to "assure that teachers, kindergarten through third grades in particular, possess the broad base of knowledge and skills ... to enable them to provide the reading approach necessary and appropriate for a wide range of ability levels among students." In addition, efforts will be undertaken to "implement the equivalent of an academic
concentration in reading/language arts ... for elementary education majors."

For existing teachers and administrators, a comprehensive plan for staff development was proposed that still would offer schools "a great deal of local control and flexibility." Under the plan, teachers would be offered multiple training options and each school would develop a staff development plan designed to achieve program goals. The report also recommended modifications to the state's teacher licensure policies that would require all North Carolina teachers in kindergarten through third grade to focus a significant portion of their continuing education on reading assessment and instruction in order to renew their teaching licenses.

• $4.7 million goes directly to the schools.

Recognizing the problems caused by the inadequacy of staff development efforts following the 1988 assessment changes, the legislature also addressed the need for staff development to implement the most recent changes. In 1997 a total of $5.2 million was appropriated to support staff development in reading and math as well as assistance to low-performing schools. Of this amount, $500,000 will be retained by the department for summer reading and math institutes, while the remaining $4.7 million goes directly to local schools.

A Delicate Balancing Act

The most difficult task facing policymakers in North Carolina and elsewhere is to understand that the strong feelings on both sides of the reading debate may not be truly reflective of the choices before them. It is not a choice between teaching reading through phonics or whole language. A good whole-language program must include phonics. And good direct- phonics instruction, while essential, is only one part of an effective reading program.

Rather, policymakers in all states should seek to walk the fine line between the more single-minded positions on either side of the reading debate to promote an approach that really is based on strong evidence of effectiveness. That means a balanced approach, and it means recognizing that "balance" will mean different things for different children and will require being able to assess accurately each child's instructional needs. It also means providing enough resources so that teachers can receive the training and support they need to understand what balance means and can attain the skills necessary to do it. And it means staying the course with this new, balanced approach for long enough to give it a chance to show results, while at the same time being prepared to make the kind of program adjustments that undoubtedly will be indicated by future re-search in the field.

The practice of teaching reading, like the practice of scientific medicine, never should cease to be, in the most positive sense of the term, a work in progress.

How to Prevent Reading Difficulties

"While science continues to discover more about how children learn to read and about how teachers and others can help them, the knowledge currently available can equip our society to promote higher levels of literacy for ... American school children."

"Beginning readers need explicit instruction [to gain] an appreciation that spoken words are made up of smaller units of sounds ..."

"Fluency should be promoted through practice with a wide variety of well-written and engaging texts at the child's own comfortable reading level."

"Because the ability to obtain meaning from print depends so strongly on the development of word recognition accuracy and reading fluency, both ... should be regularly assessed in the classroom, permitting timely and effective instructional response when difficulty or delay is apparent."

"If we have learned anything ... it is that effective teachers are able to craft a special mix of instructional ingredients for every child they work with."

"Although volunteer tutors can provide valuable practice and motivational support for children learning to read, they should not be expected either to provide primary reading instruction or to instruct children with serious reading problems."

From Preventing Reading Difficulties in Young Children, national Research Council, 1998

For further information, please contact David Denton. david.denton@sreb.org.