ABSTRACT This paper addresses instructional issues in fostering self-directed learning (an adult education concept) in the higher education classroom setting. A study of faculty members' beliefs and practices examined the nature and extent of instructor support for self-directed learning at a Canadian university. The majority (87%) of study participants did not support self-directed learning; among the small number of faculty members with attitudes that were fully supportive of self-directed learning, instructional practices did not always exemplify their apparently strong convictions about fostering students' self-directed learning. Results have implications for the ways self-directed learning is conceptualized in the adult education literature and the ways it is fostered in the higher education setting. Discussion focuses on identifying and examining the barriers to effective instructional support for self-directed learning in the modern-day university.

Self-directed learning is a process of learning in which learners function autonomously, taking responsibility for planning, initiating, and evaluating their own learning efforts. While self-directed learning is often equated with independent study and with a select group of highly motivated, experienced learners, the adult education literature suggests it is feasible and desirable to encourage self-directed learning among all learners, in a wide variety of settings.

This paper addresses, from the perspective of an adult educator, instructional issues in fostering self-directed learning in the higher education classroom setting. My perspective is reformed by a study of university instructors, a review of related literature, and reflection upon my experiences in the university setting as a self-directed learner, a facilitator of self-directed learning, and an instructional developer. I speak as an advocate for self-directed learning who seeks a critical and deep understanding of the learning process, the impact of instruction on learning, and the ways in which contextual factors influence the teaching and learning process.

After several years of practice as an adult educator, I began formal study in the field of adult and higher education. The adult education concept of self-directed learning captured my interest because it addressed, practically and theoretically, the thorny issue of learner autonomy. The following brief introduction to the adult education literature on self-directed learning is for those readers who are unfamiliar with the concept of self-directed learning as it is presented in that literature.

The origins of self-directed learning, as with many adult education principles, can be traced to John Dewey. Dewey proposed that all persons are born with an unlimited potential for growth and development; he defined education as the agency that facilitates this growth and cautioned that the teacher should be the one who guides but does not interfere with nor control the process of learning (Dewey, 1916, 1938). Contemporary adult learning theory, which values a learner-centred approach to adult education and places experience at the heart of learning, is a testament to the enduring influence of Dewey. The literature of adult education emphasises the characteristics of learners and the process of learning, and provides insight into methods that are appropriate for facilitating learning (cf. Brundage & Mackeracher, 1980). The effective teacher of adults focuses on learning topics and tasks that learners have identified as meaningful to them, and accepts that the power for growth and development is in the learner's hands. The role of the teacher in adult education is to help adults learn (thus the term: learning facilitator) The literature recommends that learning experiences for adult learners should: (a) actively involve the learners in a process aimed at resolving learner needs and concerns; (b) be characterised by supportive and collaborative instructor-learner relationships; and (c) develop learners' capacity for managing their learning.

The term 'self-directed learning' emerged in the North American literature in the mid-1970s. Tough's (1971) learning projects research had demonstrated that self-teaching was a natural process among many adults, and Malcolm Knowles built his andragogical model on the basic assumption that adult learners are self-directing (Knowles, 1975, 1980). Since that time, self-directed learning has become a prominent feature of adult education theory and practice--in fact, some educators suggest that adult education is synonymous with self-directed learning. Philip Candy (1991), in a work that is widely regarded as the most comprehensive analysis and discussion of self-directed learning to date, has constructed a conceptual framework for understanding self-directed learning as both a goal and a process which embraces four distinct phenomena: personal autonomy, self-management, learner control, and autodidaxy. Stephen Brookfield (1986) has been one of the most articulate critics of self-directed learning, warning of the dangers of an orthodoxy of self-directed learning. Yet he continues to be one of its most ardent supporters, arguing that self-directed learning honours both humanistic and critical traditions in adult education, and allows adults to achieve autonomy in and through learning (Brookfield, 1993).

In my view, the clear focus on student learning found in the field of adult education offers a valuable and distinct perspective to educators who are committed to providing effective instruction in the university setting--especially those who define effective instruction in terms of the quality of student learning. More specifically, self-directed learning's emphasis on personal autonomy, personal responsibility, and personal growth embodies some of the most fundamental principles of higher education. The common invocation that, 'universities must produce graduates who are self-directed learners' and the frequently-overheard wish of professors that, 'students should take responsibility for their own learning' lend credence to the idea that self-directed learning is suited to higher education. It is, therefore, rather surprising to discover that the concept of self-directed learning, so prominent in adult education, has been virtually ignored in the North American literature on university teaching. However, upon a closer look, one finds that some compelling arguments have been made for approaching teaching and learning in higher education from an adult education perspective. Halpern (1994), for example, reminds us that the university experience is a critical stage in adult development. Boud (1988) points out that higher education and adult education share a common interest in the goal of developing student responsibility and autonomy in learning. Similarly, Knapper & Cropley (1991) suggest that principles of adult and lifelong learning should be adopted within institutions of higher education, because students will continue to use lifelong learning skills long after graduation. Brookfield (1990) and Cranton (1989,1992) have clearly demonstrated that approaches to teaching adult learners are fully applicable in the university or college setting. Lam (1985) studied the learning experiences of university and community college students and found that the majority of learners expressed a desire for more partnership in the planning, organising, delivering, and evaluating of courses, as is advocated in the higher education literature. Finally, a specific intersection point for the adult education perspective and the higher education setting can be found in the philosophy and practice of problem-based learning in the professional schools of universities (Boud & Feletti, 1991; Woods, 1994). In fact, an underlying assumption of many persons who promote problem-based learning is that it encourages self-directed learning in students (Ryan, 1993).

Interestingly, the idea of self-directed learning is currently enjoying some popularity among those who promote change in higher education. Self-directed learning does have the potential to engage faculty with widely divergent views about education in efforts to improve educational processes. As I have noted, self-directed learning reflects the traditionally high value that universities have placed on encouraging students to take responsibility for their learning, and also reflects the liberal education value of lifelong learning--a value of particular import among those faculty members who are eager to develop employable skills in their students (a capacity for lifelong learning is a valuable asset in a rapidly-changing, employment-scarce world). Much of the current attraction of self-directed learning is clearly related to the resource crunch that defines the contemporary context for higher education. Self-directed learning seems to promise a reasonable solution to the immediate and very real problem of providing high quality educational experiences with less demand on public resources. Finally, a greater emphasis on learning, particularly student-directed learning, satisfies those faculty members who feel that undue attention has been given to the teaching side of the educational process. The idea of self-directed learning has special appeal for educators who expect students to be actively engaged in all aspects of the teaching and learning process and who rail against the notion of students as consumers of higher education.

To inquire further into the nature of support for self-directed learning in higher education, I conducted a study of faculty members' instructional beliefs and practices. My purpose was to examine whether university instructors foster self-directed learning among their students. Do instructors want learners to plan their own learning? What do instructors say about self-directed learning? What do instructors do in their classrooms? My expectation was that the results of this study would lead to a deeper understanding of self-directed learning and inform my educational practice in the field of instructional development and adult education.

Research model. The first step in the research process was the definition of a model describing the conditions of self-directed learning within a group setting (see Fig. 1). The four major categories of the model, identified through a review of the literature, are Structure, Climate, Learner Engagement, and Learner Competencies. Key indicators within each of the categories describe the specific learning conditions that are present in a self-directed learning experience. In terms of Candy's (1991) conceptual framework for the various meanings of the term self-direction, this study considered self-direction as the 'method' of education, with the emphasis on learner-controlled instruction. I assumed that an instructor must provide certain instructional structures and processes to support a stated appreciation for the value of self-directed learning. I also assumed that if learners are to assume full responsibility for their learning in a self-directed fashion, they must be involved in the planning of learning. Thus, the conditions described by the indicators in the Structure category are essential conditions for self-directed learning; if these conditions are not at least partially met, the experience cannot be labelled as a self-directed one. The conditions described by the indicators in the other three categories are considered facilitative and supportive of self-directed learning, but are insufficient conditions for classifying an experience as self-directed.

Subjects and procedure. The research design consisted of two phases. The first phase focused on instructional beliefs; its purpose was to determine the degree of attitudinal support for self-directed learning among a sample of university instructors. I constructed a 36-item Likert-type questionnaire (Orientation Survey) to elicit instructor opinions and attitudes about students, the teaching/learning process, and the role of the instructor. It was distributed to all full-time faculty members (305 persons) at a Canadian university. Some 87% of the total 139 respondents reported instructional beliefs, values, and expectations that did not support self-directed learning.

The 18 instructors (13%) whose attitudes were fully supportive of self-directed learning became the study group for the second phase of the study, which focused on instructional practices. All instructors in the study group were full-time faculty members. The mean number of years' teaching experience was 15; 87% taught primarily at the undergraduate level.

A previous study by Gorham (1985), which included analysis of observed as well as reported teaching practices among educators of adults and pre-adults, found little evidence in practice of teachers' reports that they taught adults in accordance with adult learning principles. To assess the extent to which the study group's stated support for self-directed learning was put into practice in the classroom, I used four approaches to collecting data about instructional practice:

 1. The study group completed a second survey, in which they
    described instructional practice. The Practice Survey
    consisted of 79 statements describing alternative forms of
    practice for each stage in the usual process of instructional
    design. Respondents were asked to indicate any statement that
    described their usual approach to instructional planning.

 2. I observed the study group in instructional practice within
    the classroom setting.

 3. I collected and reviewed course materials used by the study
    group to describe course structure, methods, and expectations
    to students.

 4. In a brief, unstructured way I interviewed individuals in the
    study group about their approaches to teaching.

Data analysis. I used the Statistical Package for the Social Sciences (SPSS-X) to analyse data from the Orientation and Practice Surveys, and classified these data according to the four categories of the research model: Structure, Climate, Learner Engagement, and Learner Competencies. I analysed data collected through classroom observation and course materials review by first classifying each record according to the indicators of the research model, then placing each record on a continuum that ranged from (-)--'Instructor behaviour that is not indicative of self-directed learning'--to (+)--'Instructor behaviour that is strongly indicative of self-directed learning'--and finally calculating the frequency of behaviour at each point of the 3-point continuum for each indicator.

Limitations. The study was complicated by several factors. First, since all participants in the study were, in effect, volunteers, and were from a single site, my conclusions may not be generalisable. Nonetheless, sampling procedures can be considered adequate for the practical purpose of describing the nature of instructor support for self-directed learning on a typical Canadian university campus.

Second, following on the argument that self-directed learning is a worthwhile approach to education for adults, this research looked at self-directed learning in the university setting--assuming that undergraduate and graduate students are young adults who, like all other adults, are engaged in an ongoing process of maturation and development. University instructors' perceptions of their students' adult status can have a significant impact on their approach to teaching. It was difficult, though not impossible, for this study to acknowledge the ways that instructors might mould teaching to the perceived stage of development of their students, and to extricate from this complex situation the degree of instructor support for self-directed learning in their students.

A final complication relates to contexts; most teachers adapt teaching to situational variables such as class size, subject area, length of course or of class session, background knowledge of students, etc. This confounded the task of collecting data. I intended to capture the essence of participants' teaching, and to elicit information about typical classroom teaching methods. The study used multiple and varied data collection methods so that a true picture of participants' teaching could emerge from the data. Nonetheless, instructors were sometimes asked to make general statements about their teaching when, in reality, they taught in specific ways in specific situations.

Attitudinal support. Only 13% of this sample of university faculty members reported instructional beliefs, values and expectations that were supportive of self-directed learning. Informal interviews with these faculty members (i.e. instructors who indicated attitudinal support for self-directed learning) provided additional insight into their attitudes and perceptions about self-directed learning. All the instructors described an approach to instruction that was learner-centred, and all suggested that self-directed learning was desirable, though not all saw it as the main goal of their teaching. They considered the self-directed approach to be unconventional, difficult to enact, but worthwhile in a university. Most of the instructors saw themselves as different from other university instructors in their approach to teaching. These instructors were highly committed to the teaching aspect of their professional role. They indicated that they tried to foster self-directed learning in all teaching situations, but did not expect to use the same approach in every situation nor to see the same outcome in terms of students' self-directed behaviour in every situation. In addition to the difficulties they faced in adapting the approach to suit the students they taught, they also mentioned constraints related to the expectations of colleagues and the demands of the university--particularly surrounding student assessment procedures. Instructors varied from one another in the efforts they made to fit their rather unconventional approach into the conventional university setting.

Support in practice. A look at the instructional practices of the study group (18 faculty members with attitudes fully supportive of self-directed learning) shows the various actions they took to foster self-directed learning in their students. A snapshot of instructional practice is presented here in terms of the four categories of the research model: Structure, Climate, Learner Engagement and Learner Competencies.

Structure. Most instructors provided a course structure that was flexible enough to meet a variety of student needs, though primarily predetermined rather than emergent. Instructors first considered general student characteristics and ways that previous students have responded to course format, and then planned the learning experience accordingly. However, instructors typically built in a series of 'check points', where the specific needs of individual students could be considered and the course requirements and expectations could be modified, if necessary. These instructors allowed individual students to direct specific aspects of the learning experience, but did not encourage the student body as a whole to direct class learning as a whole. There was more evidence of instructor flexibility and student direction in daily classroom activities than in the formal course structure as defined and communicated by written course materials. Instructors were more likely to define structure at the beginning of a course and more inclined to let students direct the learning experience as it came to an end.

Climate. Most instructors actively expressed appreciation and support for their students and tried to foster a positive learning environment. In addition, they emphasised co-operative rather than competitive relationships among students, and their actions tended to close, rather than widen, the gap between instructor and students. Student-instructor collaboration did not extend to include the practice of collaborative decision-making concerning course structure or student evaluation. Instructors were less likely to encourage group work for those activities that were to be evaluated than for other general learning activities.

Learner engagement. Instructors worked hard to design activities intended to engage students in learning. Instructional methods demanded active student participation in the class; most instructors placed the responsibility for running some, if not all, class sessions firmly in the hands of the students. Instructors fought an apparently natural tendency for the instructor to be the centre of attention, often taking deliberate action to put the focus on the students, rather than themselves. The fact that the instructors found such action necessary suggests that many students were either unaccustomed or resistant to the expectations of the instructor.

Courses were more learner-centred than one typically finds in higher education. The courses were subject-based, but the instructors structured them in ways that were intended to make the subject matter more meaningful to students. For the most part, it was the instructors who structured the learning, while trying to keep the students' best interests in mind. Less evident was practice that expected students to select and focus on what was most valuable for them to learn.

Learner competencies. It was difficult to identify instructional practices intended to develop student competency in learning. Clearly, most instructors did not expect students to improve their capacity for managing learning by actually planning the full course of learning, but many instructors did use independent student projects as a means of encouraging some student planning of learning experiences. Instructors, however, often specified their defined guidelines for structure and their expectations of outcome for student projects, and little leeway was left for students to plan, manage, and present a project according to guidelines that they developed.

Instructional practices did emphasise the development of thinking and problem-solving skills. And many of the course materials presented a rationale for using instructional strategies which might help students develop critical awareness of how process affects outcome. There was moderate use of student journals to encourage student reflection on personal meaning in course material. Finally, most instructors engaged in end-of-course practices--reflective activities, assessment procedures, response papers--intended to draw students' attention to the learning process that had just taken place.

A similar study conducted by Grabove (1994) at a Canadian community college found that only 8% of the study sample were supportive of self-directed learning, and the results of this study of faculty members at a Canadian university are congruent with those findings. The majority of study participants did not provide instructional support for self-directed learning. Among the small number of faculty members with attitudes fully supportive of self-directed learning, instructional practices generally provided only partial support for self-directed learning. The instructors took action to build a supportive and collaborative climate for learners, and to engage learners fully in active learning experiences. They were less successful in providing opportunities for learners to plan and direct their own learning, and in developing students' capacity for self-directed learning.

Unfortunately, this study did not include an opportunity to explore with participants the reasons why their instructional practice did not always exemplify their apparently strong convictions about fostering students' self-directed learning. I offer the following possible explanations:

1. Instructors were simply unaware of the ways in which theft practices did not support their beliefs. Instructors who tend, out of habit, to structure all educational experiences in the same way, can become blind to the influence of that structure on student learning, and neglect to align their teaching practice with their attitudes about learning.
2. Instructors lacked the instructional skills they needed to implement their beliefs effectively. Many of the instructors shied away from discussing teaching techniques (too 'gimmicky'), believed that their purposes and good intentions would be communicated by their general attitude and approach with students, and were unable to describe with any specificity what they did to enhance self-directed learning. Instructors who disavow the need to devise an explicit strategy for making their educational vision a reality are less able to identify and develop the instructional skills they need.
3. Instructors adapted instructional practices to suit the characteristics of their students (and the demands of the university setting). Student characteristics--level of maturity, expectations, background, degree of interest in learning--need to be considered in developing an appropriate instructional strategy. Some instructors recognised that, as Grow (1991) and Cranton (1992) have proposed, their teaching methods must be matched to the learners' stage of self-direction. Adapting instructional practices to address the real and immediate needs of actual students is a flexible response to an instructional situation, and need not imply that an instructor is without an underlying rationale for action or without commitment to self-directed learning, in this case. (Specific institutional characteristics, such as class sizes and obligations to report course plans and grades, also shaped instructional practice.)
4. Instructors had a different conception of self-directed learning from that defined in the research model, a difference which became apparent in practice. Some instructors defined self-directed learning as self-sufficiency in learning. Self-sufficiency is a personal attribute and is of some consequence in establishing instructional practice: the self-sufficient learner needs little support or assistance, works effectively in isolation, and adapts easily to the demands of quite rigorous courses. An instructor could give a self-sufficient learner choices about many small aspects of the course, yet not expect the learner to set the direction for the course of learning, or to evaluate the results. This last point highlights the essential difference between self-direction as self-sufficiency and self-direction as learner control over instruction through opportunities to make choices and participate in educational decisions.
5. Instructors did not really believe in self-directed learning, and this lack of commitment became apparent in their instructional practices. Self-directed learning includes, by definition, some aspects of learner control over the instructional process. Instructors who do not share important course decisions with students, are indicating through their actions that they are not committed to self-directed learning. These instructors may expect students to take responsibility for their learning, yet at the same time limit the students' ability to do so by retaining their institutionalised right to hold final decision-making power in their own hands.
6. Instructors' visions of self-directed learning were limited by conventions of the university setting. Instructors treated many conditions of the setting largely as assumptions when defining instructional practice. For example, most assumed that the responsibility for course decisions belonged to them, not the learners, because the institution treated these decisions as the professional responsibility of faculty members. While at first glance this is a pragmatic and acceptable response to the particulars of a situation, a closer look uncovers the danger in allowing the limits of the instructional setting to become the rationale for the instructional approach. Self-directed learning is presented in the adult education literature as a means for fully engaging learners in meaningful learning experiences and meaningful dialogue concerning those experiences. While a self-directed approach may have to be adapted to specific situations, it is foolhardy to abandon it simply because it does not reflect the conventions of the university setting. The possibilities for alteration of the setting should be considered as well, if the long-term needs of learners as learners are to be met.

One way to respond to this study is to conclude that instructional support for self-directed learning is inappropriate and/or unnecessary in higher education. In fact, 87% of study participants communicated this belief to me through their responses to the Orientation Survey. Yet I cannot ignore those 18 instructors whose responses reflected a strong commit-merit to a self-directed approach, and whose practice in many ways demonstrated their dedication to fostering self-directed learning in their students. Their experiences--and my own, when added to the literature that supports self-directed learning in many other contexts, suggest to me that the self-directed approach is a valid one for university teaching and learning in higher education and that more strenuous attempts to promote and study self-directed learning in higher education are warranted.

Another form of response, then, is simply to look for practical ways to promote instructional support for self-directed learning. The study has several immediate implications for instructional development. Instructors who are keen on the idea of self-directed learning, but inexperienced, could be given practical guidelines and training in specific instructional techniques for implementing the approach in the classroom. Instructors who are already actively engaged in efforts to foster self-directed learning could be provided with structured opportunities to interact with one another; they could, for example, discuss which aspects of the approach they feel most comfortable with and which aspects are the most difficult, and share tips on effective classroom strategies. Finally, the fact that only a very small percentage of faculty members in the university reported attitudes supportive of self-directed learning suggests another type of intervention: campus-wide awareness sessions that present information about self-directed learning and opportunities to meet instructors who are committed to the approach.

A third kind of response is to identify and examine more deeply the barriers to effective instructional support of self-directed learning in the modern-day university. Is self-directed learning viable in higher education? In this last section of my discussion, I will explore some issues in self-directed learning and higher education in an effort to answer this question.

Self-directed learning raises questions about who has authority to define what is to be learned and how it is to be learned in the university. When learners are asked to consider what is worth knowing, and invited to participate in decisions about the design and assessment of learning experiences, Pandora's box is opened. Some faculty, with the right combination of personal characteristics (especially commitment) and situational circumstances, are able to create learning spaces in which those questions can be openly raised and productively discussed. Many, many more will find that they are not able to do this.

As this study has demonstrated, 'knowledge . . . is socially constructed and . . . accordingly learning is a social process . . . Self-direction does not necessarily imply solitary learning' (Candy, 1991, p. 367). One hopes that self-directed students will build sound, authentic, and meaningful connections between themselves and the stuff they study and the persons they study with; self-directed learning offers students an opportunity to act autonomously to make choices about what and who they will connect with and care for. Connection and autonomy are both extremely important in self-directed learning--in fact, I would say they are essential. Yet too little attention is given in the adult education literature to the problem of how to foster, at the same time, autonomy through independence and connection through relation. I propose that in effective self-directed learning, connection and autonomy interact dialectically and the facilitator of self-directed learning plays a tremendous role in making this dialectic possible--through caring. My argument that caring allows relationships of connection and autonomy to thrive is based on Noddings's (1984) description of an ethic of care. Noddings states that an ethic of caring is characterised by receptivity and responsiveness; we keep ourselves open to the unique needs and interests of those we care for so that we may respond to them and thereby continue our caring relationship with them. Caring is a contextualised, constructed process, which allows persons to become themselves through connections with others. From a perspective of care, facilitating self-directed learning is less a matter of turning students loose to pursue their goals, and more a matter of supporting students while they struggle to construct a self in relation to a discipline or profession.

It is my contention that, within institutions of higher education today, many instructors lack first, the autonomy (real or felt) they themselves must have if they are to offer autonomy to their students, and second, the skills and attitudes they need to build and maintain caring connections with students. Individual instructors may or may not have the capacity for caring, but more importantly, these skills and attitudes of caring are not highly valued in our universities, which have been founded on an ethic of rights, principles and rules. I conclude that our inattention to the importance of care in the process of fostering self-directed learning, and the fact that an ethic of care is simply not the dominant ethic underlying decisions, programmes, relationships within higher education, are powerful barriers to self-directed learning in higher education. Self-directed learning would have greater viability if caring was recognised and supported in our institutions. If an institution does not reflect the ethic of caring that fosters self-directed learning, it is difficult--perhaps impossible--to foster self-directed learning within that institution.

The rewards given to learning and scholarship in higher education illustrate the ethic that does govern most actions in that setting. In universities, it is accepted practice to reward scholarly work with honour. Universities have established a kind of academic heroism, which plays an important role in nourishing scholars' motivation for scholarly activity--particularly necessary when scholars are separated from those in the 'real world' who stand to benefit from their work, as is frequently the case with higher education research and scholarship. This ethic has been extended to the realm of teaching, by giving teaching awards to instructors to honour their work with students. However, from the perspective of an ethic of care, awards for teaching are an anachronism. In caring teaching and learning relationships, it is the response of the cared-for (learner) that acts as feedback for instructors and encourages them to continue their teaching, and it is learners' sense of being cared for that encourages them to continue their study. It is worth noting that this response feedback mechanism works only if connections between those in the relationship are fostered and maintained, and that it is threatened as classes grow larger and teacher--student connections are broken. Strangely enough, self-directed learning invites instructors to establish caring relationships with students, grounded as it is in a dialectic of autonomy and connection.

As some of the barriers to self-directed learning become apparent, it is worth reconsidering whether it is truly necessary to raise with students the difficult questions I referred to earlier--questions about what is worth learning and who decides what is worth learning in the university. I wish to make two points in this regard. First, these questions are at the very heart of inquiry in a post-modern age. If we fail to address these questions with learners, we have failed to engage them in the process of inquiry that marks true scholarship. We have patronised them, given them a false sense of what it means to be educated. They will have gained information rather than knowledge, and even a substantial increase in information is a meagre substitute for growth in knowledge.

Second, no notion of inquiry is complete without attention to how the value of the knowledge gained through inquiry is to be assessed. Barnett (1994) has articulated the limits of competence as a guiding rationale for assessing the worth of learning in higher education. Noddings (1984) proposes that learners must be able to construct for themselves an ethical ideal as a standard against which they may judge themselves. When learners engage in discussion about what knowledge is needed, they are constructing a standard against which their own knowledge may be assessed. The facilitator of self-directed learning is in the unique position of being able to establish a relationship of care with learners, a relationship of a type that permits them to help learners construct an ethical ideal, an opening in which the autonomous person/learner may begin to construct an image of their best self in relation to the study of a discipline, or the practice of a profession. The instructors in my study appeared to be structuring and influencing the learning situation more than I had expected to see in instructors who valued student autonomy in self-directed learning. I now wonder whether they were showing learners what an ethical ideal might look like in that discipline, and modeling for learners how an ideal shapes further learning--intending to help learners in the process of constructing their own ideal.

The crux of the problem, I believe, lies in the fact that self-directed learning challenges the right of universities to set the ethical ideal. Universities, originally given the mandate to honour and reward learning, have over a period of time come to interpret that mandate as meaning that they can and should define what is worth learning in today's world. This is a distortion of their original need to set themselves apart from other cultural institutions and give themselves the necessary authority to fulfill their role of honouring and rewarding scholarship. Universities do have an important part to play in the construction of the all-important ethical ideal--they can be the setting in which teachers work with students in dialogue, helping them construct a personal ideal. Universities cannot, however, define the ideal institutionally for their members. Noddings (1984) states clearly that there are no short-cuts to the development of an ethical ideal, and that enforcing an ideal through institutionalised principles and rules is a poor substitute for encouraging individuals to meet the standards provided by a personally-constructed ethical ideal. When universities define what is worth learning, they act as institutional arbiters of what is intellectually valuable; this is a powerful barrier to self-directed learning in the university.

Let us imagine, for a moment, that we do accept that the university's first obligation is to develop, maintain, and promote standards of excellence in academic achievement and professional training. What would this imply for self-directed learning? Within such a system, self-directed learning is likely to be encouraged only if it is seen to facilitate excellence in academic achievement and professional training. It is not likely to be encouraged if it is seen to compromise standards. However, instructors with knowledge of the self-directed approach would recognise that self-directed learning does not always result in an end product that meets preconceived standards, especially standards as defined by others. The adult education model of self-directed learning concerns itself with the process of learning and the identity of the learner, and proposes that the desired result from a self-directed learning episode is growth, change and development--learning that is personally meaningful, and therefore particularly useful. University educators committed to self-directed learning might accept that the primary concerns of each approach are different and recognise that structures set up to support each approach will often be in conflict, yet still conclude that self-directed learning can co-exist with the obligations of higher education--through appropriately designed instruction. Instructors in this situation could cope with the conflicting demands by functioning as mediators between the interests of a self-directed learner and the institutional demands of higher education, trying to keep in mind their obligations toward both. As mediators, instructors might bring about a better balance between the learner and the institution, at least within a given course.

Although there is much to be said for such a vision of instructional support for self-directed learning, I am unable to act as advocate for this approach, constrained as it is by an acceptance of universities' inherent right to define what is worth knowing. In my view, a university's first obligation is to provide an environment in which instructors have the autonomy and support they need so that they can be effective mentors for students engaged in self-directed learning. Self-directed learners must be invited to actively construct their own best image of themselves in relation to their discipline or profession so that they may contribute meaningfully to that discipline or profession. The thorny issue of who has authority and autonomy to define what is to be learned and how it is to be learned cannot be ignored.

In this paper I have taken steps towards articulating a vision of self-directed learning within higher education. I continue to believe that educators who are truly committed to providing effective instruction in higher education, and particularly interested in broadening and enhancing the nature of classroom participation and discourse, must be alerted to the benefits of self-directed learning and supported in their efforts to encourage self-directed learning. The only way to ensure the viability of self-directed learning in higher education is to continue to invite self-directed learning and respond with care to the challenge of self-directed learners.

Correspondence: Susan Wilcox, Instructional Development Centre, Queen's University, Kingston, Ontario, Canada K7L 3N6.

FIG. 1. Research model for self-directed learning. This model describes the conditions characteristic of a self-directed learning (SDL) experience that is conducted within a group setting and facilitated by an educator. The conditions described in the STRUCTURE category are essential and necessary conditions for SDL The conditions described in the Climate, Learner Engagement, and Learner Competencies categories are facilitative of, but not essential to, SDL.

I. Structure

* Flexible emergent plan, responsive to learner needs

* Options for learner, i.e. learner choices concerning:

-- participation

-- objectives

-- content

-- methods/materials

-- pace

-- evaluation

II. Climate

* Supportive

* Collaborative

III. Learner Engagement

* Focus on problems of learners

* Learners are actively involved

IV. Learner Competencies

* Awareness of learning process

* Reflect on personal meaning

* Develop learning skills/strategies

BARNETT, R. (1994) The Limits of Competence: knowledge, higher
     education, and society (Buckingham, Open University Press and
     The Society for Research into Higher Education).

BOUD, D. (Ed.) (1988) Developing Student Autonomy in Learning, 2nd
     edn (London, Kogan Page).

BOUD, D. & FELETTI, G. (Eds) (1991) The Challenge of Problem-based
     Learning (London, Kogan Page).

BROOKFIELD, S. (1986) Understanding and Facilitating Adult
     Learning (San Francisco, CA, Jossey-Bass).

BROOKFIELD, S. (1990) The Skillful Teacher (San Francisco, CA,
     Jossey-Bass).

BROOKFIELD, S. (1993) Self-directed learning, political clarity,
     and the critical practice of adult education, Adult Education
     Quarterly, 43, pp. 227 242.

BRUNDAGE, D. & MACKERACHER, D. (1980) Adult Learning Principles
     and their Application to Program Planning (Toronto, Ontario
     Institute for Studies in Education).

CANDY, P.C. (1991) Self-direction for Lifelong Learning (San
     Francisco, CA, Jossey-Bass).

CRANTON, P.A. (1989) Planning Instruction for Adult Learners
     (Toronto, Wall & Thompson).

CRANTON, P.A. (1992) Working with Adult Learners (Toronto, Wall
     & Emerson).

DEWEY, J. (1916) Education and Democracy (New York, Macmillan).

DEWEY, J. (1938) Experience and Education (London, Collier
     Macmillan).

GORHAM, J. (1985) Differences between teaching adults and
     pre-adults: a closer look, Adult Education Quarterly, 35, pp.
     194-209.

GRABOVE, V. (1994) Instructional support for student-directed
     learning in an Ontario community college, paper presented at
     the Annual Meeting of the Canadian Society for the Study of
     Higher Education in Calgary, Alberta.

GROW, G.O. (1991) Teaching learners to be self-directed, Adult
     Education Quarterly, 41, pp. 125-149.

HALPERN, D. (Ed.) (1994) Changing College Classrooms (San
     Francisco, CA, Jossey-Bass).

KNAPPER, C.K. & CROPLEY, A.J. (1991) Lifelong Learning and Higher
     Education, 2nd edn (London, Kogan Page).

KNOWLES, M. (1975) Self-directed Learning (New York, Association
     Press).

KNOWLES, M. (1980) The Modern Practice of Adult Education (New
     York, Association Press).

LAM, Y.L. JACK (1985) Exploring the principles of andragogy: some
     comparisons of university and community college learning
     experiences, Canadian Journal of Higher Education, 15, pp.
     39-52.

NODDINGS, N. (1984) Caring: a feminine approach to ethics and
     moral education (Berkeley, CA, University of California
     Press).

RYAN, G. (1993) Student perceptions about self-directed learning
     in a professional course implementing problem-based learning,
     Studies in Higher Education, 18, pp. 53-63.

TOUGH, A. (1971) The Adult's Learning Projects (Toronto, Ontario
     Institute for Studies in Education).

WOODS, D. (1994) How to Gain the Most from Problem Based Learning
     (Hamilton, McMaster University).

~~~~~~~~

By SUSAN WILCOX

Queen's University, Ontario, Canada