On-line version ISSN 2072-8050
Herv. teol. stud. vol.65 n.1 Cape Town Jan. 2009
Department of Theology, York St John University, United Kingdom
A sample of 404 Anglicans from a variety of church traditions within the Church of England was asked if they could imagine themselves into a healing story from Mark 9:14-29 by identifying with one of the characters in it. Around 65% could do so ('imaginers') and 35% could not. The likelihood of being an imaginer was higher among (i) women than among men, (ii) those who preferred intuition to sensing or feeling to thinking, and (iii) those who were most charismatically active. Readers with intuition as their dominant function were most likely to be imaginers, while those with thinking as their dominant function were least likely to be so.
Keywords: Church of England; Anglican; Mark 9:14-29; psychological type preferences; religious preferences
Ignatius of Loyola, founder of the Jesuits, is also remembered for his spiritual exercises, some of which required disciples to imagine themselves in a particular place or event described in the gospels (Ganss 1998; O'Malley 1993). The objective of the exercises was to allow the exercitant to discern the will of God by engaging imaginatively with scripture or with other religious ideas. This way of reading falls under the wider category of lectio divina, a contemplative approach to scripture that was symptomatic of biblical hermeneutics in the premodern period (Thiselton 1992:142). The spiritual exercises developed this sort of meditation by expanding the use of the reader's imagination (Barth 2000; Ferlita 1997).
Imaginative engagement with scripture in this way has a long tradition within the Catholic Church, but in the latter half of the twentieth century it began to be used more widely as part of a general increase in interest in Christian spirituality. It has formed the basis of popular guides to meditation and prayer (Hughes 1996; Lonsdale 2000), including some produced by evangelical writers such as Joyce Hugget (1986; 1989; 1990). Retreats based on the spiritual exercises or similar sorts of scripture-based contemplation are now commonplace in the programmes of retreat houses. It is not uncommon during such programmes for participants to be asked to 'imagine themselves' into a biblical passage, often by identifying with one of the characters in the story. Those who have led such meditations in churches or on retreats will know that they can evoke mixed reactions. Some participants find the exercises easy to do and emerge from the session energised, excited or challenged. Others find it virtually impossible to understand what is being asked of them, and can emerge from the session bemused, irritated or angry.
A number of writers have explored the psychology of the imaginative techniques associated with the spiritual exercises. William Meissner's (1999) approach to the spiritual exercises is mainly psychoanalytical, and stems from his understanding of the psychological background of Ignatius himself. Kenneth Becker (2001), on the other hand, explores the interaction of Carl Jung with the spiritual exercises. Jung lectured on the exercises in 1939/40 and saw in them an analogy with his concept of achieving individuation through the process of active imagination (Becker 2001:29-39). This direct link between Jung and the spiritual exercises stands alongside a more indirect link that arises from those who have used Jung's theory of psychological type to predict general religious and spiritual preferences (Faucett & Faucett 1987; Johnson 1995; Keating 1987; Michael & Norrisey 1991; Moore 1988; Ware, Knapp & Schwarzin 1989).
The theory of psychological type posited by Jung (1921, 1971) suggested that psychological functioning is governed by the processes of perceiving (by which individuals take in information) and judging (by which they evaluate information and make decisions). Each of these processes can operate in different ways: Perceiving may rely on the senses or on the imagination; judging may rely on thinking or feeling. Although individuals can perform each process in both ways, Jung argued that most tend to have a preferred way of perceiving and a preferred way of judging. For perceiving this is either sensing or intuition, while for judging this is either feeling or thinking. Jung also argued that the processes of perceiving and judging may operate in the interior world of the individual (introversion) or the outer world of interactions with others (extraversion). Preferences in the two psychological functions of perceiving and judging, and preference for the world in which they operate, lead to Jung's eight-fold typology.
This typology was extended by Myers and Myers (1980) who added a fourth preference set based on attitude toward the outer world (judging versus perceiving), which indicates whether individuals prefer to operate their judging or their perceiving process in the outer world. Thus a sixteen-fold typology, based on four preferences, is generally used to describe psychological type preferences today:
Extraversion versus introversion (EI)
Extraverts are energised by the outer world of relationships and objects: they tend to prefer action or discussion to reflection or solitude. Introverts prefer to concentrate on the inner world of ideas: they are energised by reflection and by solitude. Extraverts will often 'think out loud' while introverts may think through things by themselves.
Sensing versus intuition (SN)
Sensing types are comfortable with familiar things perceived through the senses; intuitives thrive on new information and on imagined possibilities. Those who prefer sensing look at specific parts and pieces, rather than patterns and relationships. They are good observers and use their senses well to acquire details of their surroundings. Intuitives prefer ideas to facts and they enjoy linking information into bigger patterns of challenge and possibility.
Thinking versus feeling (TF)
Thinking types prefer to make objective decisions and impartial judgements based on careful analysis and the application of rational principles. Feeling types prefer to make subjective decisions based on personal values and standards. They work for harmony, even at the expense of fairness. Feelers find it easy to empathise with other people, whereas thinkers find it harder to gauge what other people are feeling or thinking.
Judging versus perceiving (JP)
Those who operate their judging function in the outer world tend to prefer an orderly, planned lifestyle in which decisions are made carefully and in advance of deadlines. Those who operate their perceiving function in the outer world tend to prefer a more flexible approach to life associated with open-ended decisions that may change with circumstances.
Psychological type theory also includes the notion of type dynamics (Bayne 1997; Myers & McCaulley 1985; Quenk 1993, 1999), which suggests that one of the four possible preferences in the perceiving and judging processes (S, N, T or F) is dominant (Myers & Myers 1980:1012). It is the individual's preferred function that is directed toward his or her preferred world: the inner world for introverts and the external world for extraverts.
There are several different instruments available that have operationalised psychological type theory such as the Myers-Briggs Type Indicator (MBTI), the Keirsey Temperament Sorter (KTS) and the Francis Psychological Type Scales (FPTS). The MBTI is the longest questionnaire and its use requires workshop sessions and training (Myers 2006). The KTS (Keirsey 1998; Keirsey & Bates 1978) and FPTS (Francis 2005) measure the same four dimensions and can be completed by individuals working alone. The KTS was produced by Keirsey and Bates (1978) and developed by Keirsey (1998) in order to provide an accessible operationalisation of psychological type theory. Internal reliability has been assessed using Cronbach's alpha coefficient (Cronbach 1951) from a range of studies. For example, Waskel and Coleman (1991) in a study of 331 university students reported values of .74 (EI), .89 (SN), .87 (TF) and .88 (JP). Equivalent values among 367 university students in the UK were .68 (EI), .73 (SN), .74 (TF) .82 (JP) (Fearn, Francis & Wilcox 2001). These values are above the minimum values generally recognised as indicating internal reliable scales (DeVellis 2003; Kline 2000). The KTS has also been shown to demonstrate concurrent reliability with other measures of psychological type such as the MBTI (Francis, Robbins & Craig 2005; Quinn, Lewis & Fischer 1992; Tucker & Gillespie 1993).
Psychological type and religious preferences
A number of writers have suggested that preferences in spirituality and biblical interpretation may be linked to preferences in the perceiving and judging processes (Francis 2001). Those who prefer sensing to intuition seem to be more conservative in their religious beliefs (Francis & Jones 1998; Francis & Ross 1997; Village 2005b) and tend to make a clearer separation of the sacred from the secular (Ross, Weiss & Jackson 1996). Charismatic preferences seem to be more closely linked to the judging than to the perceiving process, though whether they are associated with a preference for thinking or feeling is unclear (Francis & Jones 1997; Village 2005b). The links between psychological type and religious expression prompted the theory that psychological type may be one factor that influences the way that people interpret scripture or respond to preaching (Francis 2003). Francis (1997) and Francis and Atkins (2000; 2001; 2002) used this theory to produce short homilies based on gospel lectionary readings that were intended to reflect preferences for either sensing, intuition, feeling or thinking. Village and Francis (2005) showed that lay Anglicans did indeed show a preference for interpretations that matched their preferences for sensing verses intuition or feeling versus thinking.
Psychological type could influence the ability of people to imagine themselves into a biblical story, though the effects might operate in several ways. Such an exercise is essentially one of introversion, and might be inherently difficult for extraverts. Those who prefer intuition over sensing might find it easier to deal with an exercise that requires use of the imagination rather than sensing the immediate environment. On the other hand, some meditations deliberately rely on 'imagined senses' and this sort of approach might appeal to sensing types. (A good example of combining imagination and senses is found in the fifth contemplation of the nativity in the Ignatian spiritual exercises. This requires imagining the scene through the senses of sight, sound, smell and taste). The ability to empathise that is associated with a preference for feeling over thinking might make it easier for an exercitant to take on the persona of a character in a story.
This study explores the idea that psychological type preferences might be an important factor shaping the ability to imagine oneself as a character in a bible story. Responses to a test passage were used to identify those who claimed that they were able to do so and those who were not. Multivariate analysis was used to test for associations between being able to imagine oneself as a character in a story and psychological type preferences, after controlling for other factors related to psychological type and biblical interpretation. It was predicted that being able to identify with a character in this way would be associated with introversion rather than extraversion, intuition rather than sensing and feeling rather than thinking.
The data were collected from 404 Anglican churchgoers in England. Questionnaires were returned from 11 different congregations of varying size and church tradition and comprised roughly equal numbers from Anglo-Catholic, Evangelical and broad churches. Comparison of the overall sample with that recorded in the 1998 English Church Attendance Survey (Brierley 2000) showed that it spanned much of the variation found in the Church of England and contained an almost identical gender ratio (63% women). However, in the present sample there were more middle-aged people and more frequent church attendees than expected from the church at large (Village 2007).
Participants were asked to read Mark 9:1429, a story about the healing of a boy who is possessed by an evil spirit. The text was from the New Revised Standard Version, but with book, chapter and verse annotations removed. Two questions followed the story. In response to the first question, 'Have you heard this story before?', participants could respond 'yes', 'no' or 'I'm not sure'. The latter were grouped with negative responses for analysis. The second question was: 'Some people are able to imagine themselves "into the story". They picture the scene, often by identifying with a character in the story. If you are able to imagine yourself as part of the story, who would you be?'Possible answers were 'Jesus', 'the boy', 'one of the disciples', 'the boy's father' and 'I can't imagine myself in the story'. For this analysis, responses were categorised as 'imaginers', who identified with a character, and 'non-imaginers' who did not.
Psychological type preferences were assessed using the Keirsey Temperament Sorter (KTS), the published version of which was stapled to the inside of the questionnaire. This instrument measures the same four dimensions of psychological type as the Myers-Briggs Type Inventory. The answers were scored as per the directions on the sorter: Scores for preferences within a dimension are always complementary and sum to ten for extraversion-introversion and 20 for sensing-intuition, feeling-thinking and judging-perceiving. Individuals were assigned types based on the highest score in each dimension. Those with equal scores in any dimension were treated as missing data. Dominant process was assigned using standard procedures (Bayne 1997:50; Myers & Myers 1980:1516).
Numbers of each type were compared with imagining using a 2 x 4 contingency table.
The questionnaire also assessed sex (0 = male, 1 = female), age (to the nearest decade), general educational experience (0 = no formal qualifications to 4 = postgraduate qualification), belief about the Bible and charismatic practice. The bible scale (Village 2005a) was a 12 item Likert scale that measured conservative versus liberal belief about the Bible (high score = conservative belief, Cronbach's α = .91). The five item charismatic scale (Village 2005b) assessed the frequency of charismatic practices such as speaking in tongues, speaking words of prophecy and laying hands on people for healing (high score = frequent practice, α = .74). The dependent variable, imagining, was coded 0 for those who could not imagine themselves into the story and 1 for those who could.
Multivariate analysis used binary logistic regression to identify significant predictors of imagining. The aim was to test for the effects of preferences after allowing for age, sex, education, bible score and charismatic score.
Initial analyses of imagining
Of 398 participants who gave valid replies, 35% could not imagine themselves as a character in the story and 65% could. Overall, 2% imagined themselves as Jesus, 3% as the boy, 26% as the father and 32% as one of the disciples. A small proportion (2%) wrote 'crowd' on the questionnaire, though this was not a choice given on the list.
There was no significant difference in imagining between people from churches of different traditions, with 60% of 92 Anglo-Catholics, 65% of 104 broad church members and 67% of 198 evangelicals classed as imaginers (χ2 = 1.5, df = 2, ns). Women were much more likely than men to be able to imagine themselves into the story (70% of 248 women versus 56% of 146 men, χ2 = 7.9, df = 1, p < .01). A large majority of those who answered the question had heard the story before (89%, n = 373). They were more likely (66%, n = 331) to imagine themselves in the story that those who had not heard it before (50%, n = 42, χ2 = 4.2, df = 1, p < .05). For continuous variables, there was no difference between imaginers and non-imaginers in average age or level of education (see Table 1). However, average charismatic scores and bible scores were significantly higher among imaginers than non-imaginers.
For psychological type preferences, there were no significant differences in the proportion of imaginers between extraverts and introverts (E 64%, n = 190 versus I 67%, n = 144, χ2 = 0.3, df = 1, ns), or between those who preferred to use their judging or perceiving functions in the outer world (J 65%, n = 355 versus P 68%, n = 25, χ2 = 0.1, df = 1, ns). Preferred intuitives were more likely to be imaginers than preferred sensors, but the difference was not quite statistically significant (N 73%, n = 67 versus S 64%, n = 312, χ2 = 2.1, df = 1, p = .1). The most striking difference was the higher frequency of imaginers among preferred feelers than among preferred thinkers (F 70%, n = 265 versus T 55%, n = 104, χ2 = 7.9, df = 1, p < .01).
Multivariate analysis indicated that being female rather than male, preferring intuition rather than sensing, preferring feeling rather than thinking and having a high charismatic score were significant predictors of imagining after allowing for other variables in the model (see Table 2).
When preferences for perceiving and judging were combined, there were four possible function pairs: SF, ST, NF, NT. Overall, imagining was most frequent among NFs and least frequent in STs (Table 3). This pattern remained when the sexes were treated separately though the differences were not statistically significant in either.
The importance of intuition for imagining also emerged from an examination of dominant function. Imagining was most frequent among dominant intuitives, intermediate among dominant sensors and feelers and lowest among dominant thinkers (see Table 4). This trend remained when the sexes were treated separately, but was statistically significant only in women.
Psychological type preferences seemed to exert some effect on imagining oneself into a biblical story, but the results were not entirely as predicted by theory. There seemed to be no relationship to introversion or extraversion, perhaps because this was a questionnaire-based study that did not actually require people to spend time in silent meditation. The strongest relationship between imagining and psychological type was in the judging process, with those who preferred feeling being more likely than those who preferred thinking to imagine themselves as a character in the story. This would accord with the feeling type propensity to relate to others and to having a developed ability to empathise. Thinking types might approach the story in a more analytical fashion that would find identification with a character more problematical.
Alongside empathy with a character, a preference for intuition was also important for the ability to imagine oneself into the story. The test variable was specifically about relating to a character in the story, and there was less emphasis on imagining other aspects, so a preference for intuition may not have been as important here as a preference for feeling. However, there was additional evidence that both intuition and feeling were important in imagining because NFs were the most likely group to imagine themselves into the story, and STs the least likely. In addition, the high frequency of imaginers among dominant intuitives also points to the importance of intuition to enable people to engage in this sort of exercise. A preference for intuition may predispose readers to the imaginative perception that is required to make the story become real in the absence of actual sensory information. A preference for feeling may further predispose readers to identify empathetically with these imagined characters.
Women seemed generally more able than men to imagine themselves into the story, and this difference was not wholly accounted for by their greater preference for feeling versus thinking compared to men. The apparent propensity of women over men to participate in these exercises could be mainly due to underlying differences in psychological type but there may be some other gender related propensity to identify with biblical characters in this way. Sex difference was only weakly significant in the final model, and more work is needed to see whether there are sex differences in imagining that are not related to psychological type preferences.
The apparently strong effect of charismatic practice on imagining remained even after allowing for the effects of psychological type. It may reflect the fact that charismatics were likely to see this story as a paradigm for the sorts of healing and exorcisms that are part of their religious experience. The link might then be an experiential one, so that it is easier to imagine oneself in a biblical story if that story has strong echoes of what happens in everyday life. This suggests that psychological predispositions are tempered by experience, something also indicated by the effect of having heard the story before. In the initial analyses, familiarity seemed to make it more likely that participants would imagine themselves into the story. However, this effect disappeared on multivariate analysis, possibly due to the fact that relatively few people in this sample were unfamiliar with the story, and most of these were men with low charismatic scores. If experience is a key influence on this sort of reading exercise, responses may vary considerably depending on the exact passage or idea being explored. Future work could look at reactions to a variety of passages that link in different ways to the experience of modern readers.
Although there is no exact way of calculating effect size for binary logistic regression models, the pseudo-R2 values suggested that the model incorporating sex, psychological type preferences and charismatic practice scores was a relatively poor predictor of the likelihood of someone imagining themselves as a character in this story. Clearly, the process of identifying with Scripture in this way is a complex business and may be influenced by a wide range of factors. However, this study suggests that a more thorough empirical investigation of Scripture and imagination would be both feasible and worthwhile.
Barth, J.R., 2000, 'Mortal beauty: Ignatius Loyola, Samuel Taylor Coleridge, and the role of imagination in religious experience', Christianity and Literature 50, 6978. [ Links ]
Bayne, R., 1997, The Myers-Briggs Type Indicator: A critical review and practical guide, Stanley Thornes, Cheltenham. [ Links ]
Becker, K.L., 2001, Unlikely companions: C G Jung on the spiritual exercises of Ignatius of Loyola, Gracewing, Leominster. [ Links ]
Brierley, P., 2000, The tide is running out, Christian Research, London. [ Links ]
Cronbach, L.J., 1951, 'Coefficient alpha and the internal structure of tests', Psychometrika 16, 297334. [ Links ]
DeVellis, R.F., 2003, Scale development: Theory and applications, 2nd edn., Sage, London. [ Links ]
Faucett, R. & Faucett, C.A., 1987, Personality and spiritual freedom: Growing in the Christian life through understanding personality type and the Myers-Briggs Type Indicator, Doubleday, New York. [ Links ]
Fearn, M., Francis, L.J. & Wilcox, C., 2001, 'Attitude towards Christianity and psychological type: A survey among religious studies students', Pastoral Psychology 49, 341348. [ Links ]
Ferlita, E., 1997, 'The road to Bethlehem is it level or winding?', Studies in the Spirituality of Jesuits 29, 123. [ Links ]
Francis, L.J., 1997, Personality type and Scripture: Exploring St Mark's Gospel, Mowbray, London. [ Links ]
Francis, L.J., 2001, 'Personality type and communicating the Gospel', Modern Believing 42, 3246. [ Links ]
Francis, L.J., 2003, 'Psychological type and biblical hermeneutics: SIFT method of preaching', Rural Theology 1, 1323. [ Links ]
Francis, L.J., 2005, Faith and psychology: Personality, religion and the individual, Darton, Longman & Todd, London. [ Links ]
Francis, L.J. & Atkins, P., 2000, Exploring Luke's Gospel: A guide to the Gospel readings in the Revised Common Lectionary, Continuum/Mowbray, London. [ Links ]
Francis, L.J. & Atkins, P., 2001, Exploring Matthew's Gospel: A guide to the Gospel readings in the Revised Common Lectionary, Continuum, London. [ Links ]
Francis, L.J. & Atkins, P., 2002, Exploring Mark's Gospel: An aid for readers and preachers using year B of the Revised Common Lectionary, Continuum/Mowbray, London. [ Links ]
Francis, L.J. & Jones, S.H., 1997, 'Personality and charismatic experience among adult Christians', Pastoral Psychology 45, 421428. [ Links ]
Francis, L.J. & Jones, S.H., 1998, Personality and Christian belief among adult churchgoers, Journal of Psychological Type 47, 511. [ Links ]
Francis, L.J., Robbins, M. & Craig, C.L., 2005, 'Two different operationalizations of psychological type: Comparing the Myers-Briggs Type Indicator and the Keirsey Temperament Sorter', in R.A. Degregorio (ed.), New developments in psychological testing, pp. 119-138, Nova Science Publishers, New York. [ Links ]
Francis, L.J. & Ross, F.J., 1997, 'The perceiving function and Christian spirituality: Distinguishing between sensing and intuition', Pastoral Sciences 16, 93103. [ Links ]
Ganss, G.E., 1998, The spiritual exercises of Saint Ignatius: A translation and commentary, Loyola Press, Chicago. [ Links ]
Huggett, J., 1986, Listening to God, Hodder & Stoughton, London. [ Links ]
Huggett, J., 1989, Open to God, Hodder & Stoughton, London. [ Links ]
Huggett, J., 1990, The smile of love, Hodder & Stoughton, London. [ Links ]
Hughes, G.W., 1996, God of surprises, Hodder & Stoughton, London. [ Links ]
Johnson, R., 1995, Your personality and your spiritual life, Monarch, Croborough. [ Links ]
Jung, C.G., 1921, Psychologische Typen, Rascher Verlag, Zurich. [ Links ]
Jung, C.G., 1971, Psychological Types: The collected works, vol. 6, Routledge and Kegan Pall, London. [ Links ]
Keating, C.J., 1987, Who we are is how we pray: Matching personality and spirituality, Twenty-Third Publications, Mystic. [ Links ]
Keirsey, D., 1998, Please understand me II: Temperament, character and intelligence, Prometheus Nemesis, Del Mart. [ Links ]
Keirsey, D. & Bates, M., 1978, Please understand me, 3rd edn., Prometheus Nemesis, Del Mar. [ Links ]
Kline, P., 2000, The handbook of psychological testing, 2nd edn., Routledge, London. [ Links ]
Lonsdale, D., 2000, Eyes to see, ears to hear: An introduction to Ignatian spirituality, rev. edn., Darton, Longman & Todd, London. [ Links ]
Meissner, W.W., 1999, To the greater glory: A psychological study of Ignatian spirituality, Marguette University Press, Milwaukee. [ Links ]
Michael, C.P. & Norrisey, M.C., 1991, Prayer and temperament: Different prayer forms for different personality types, The Open Door Inc., Charlottesville. [ Links ]
Moore, R.L. (ed.), 1988, Carl Jung and Christian spirituality, Paulist Press, Mahwah. [ Links ]
Myers, I.B., 2006, MBTI manual: A guide to the development and use of the Myers-Briggs Type Indicator, Consulting Psychologist Press, Palo Alto. [ Links ]
Myers, I.B. & McCaulley, M.H., 1985, Manual: A guide to the development and use of the Myers-Briggs Type Indicator, Consulting Psychologists Press, Palo Alto. [ Links ]
Myers, I.B. & Myers, P.B., 1980, Gifts differing, Consulting Psychologists Press, Palo Alto. [ Links ]
O'Malley, J.W., 1993, The first Jesuits, Harvard University Press, Cambridge. [ Links ]
Quenk, N.L., 1993, Beside myself: The inferior function in everyday life, Consulting Psychologist Press, Palo Alto. [ Links ]
Quenk N.L., 1999, Essentials of Myers-Briggs Type Indicator assessment, John Wiley & Sons, New York. [ Links ]
Quinn, M.T., Lewis, R.J. & Fischer, K.L., 1992, 'A cross-correlation of the Myers-Briggs and the Keirsey instruments', Journal of College Student Development 33, 279280. [ Links ]
Ross, C.F.J., Weiss, D. & Jackson, L.M., 1996, 'The relation of Jungian psychological type to religious attitudes and practices', International Journal for the Psychology of Religion 6, 263279. [ Links ]
Thiselton, A.C., 1992, New horizons in hermeneutics, HarperCollins, London. [ Links ]
Tucker, I.F. & Gillespie, B.V., 1993, 'Correlations among three measures of personality type', Perceptual and Motor Skills 77, 650. [ Links ]
Village, A., 2005a, 'Assessing belief about the Bible: A study among Anglican laity', Review of Religious Research 46, 243254. [ Links ]
Village, A., 2005b, 'Christian belief about the Bible and the Holy Spirit in relation to psychological type', Review of the Social Scientific Study of Religion 16, 116. [ Links ]
Village, A., 2007, The Bible and lay people: An empirical approach to ordinary hermeneutics, Ashgate Publishing, Aldershot & Burlington. [ Links ]
Village, A. & Francis, L.J., 2005, 'The relationship of psychological type preferences to biblical interpretation', Journal of Empirical Theology 18, 7489. [ Links ]
Ware, R., Knapp, C.R. & Schwarzin, H., 1989, 'Prayer form preferences of Keirsey temperaments and psychological types', Journal of Psychological Type 17, 3942. [ Links ]
Waskel, S.A. & Coleman, J., 1991, 'Correlations of the temperament types, intensity of crisis at midlife with scores on a death scale', Psychological Reports 68, 11871190. [ Links ]
York St John University
Lord Mayor's Walk
York, YO31 7EX, United Kingdom
Received: 05 Jan. 2009
Accepted: 26 Mar. 2009
Published: 04 Aug. 2009
This article is available at: http://www.hts.org.za