The Australian Quarterly, 1982, 54, 435-443.


John Ray

The issues surrounding penal policy have been arousing considerable passion among concerned people at least since the Victorian era. A rather strong impression one sometimes gets is that those who are closest to the victim tend to think that the penal system is not punitive enough, while those who identify more with the criminal think that any penal policy at all is distinctly dubious.

Given this great range of opinion, a seemingly obvious question might be how the public in general sees the matter. Is the public on the side of the reformers or of the 'reactionaries'? As there has been, throughout this century, a steady amelioration of prison conditions and a steadily reducing use of custodial sentences, it might well be held that a liberal climate of public opinion has in fact gradually forced abandonment of the more punitive practices of the past.

The most systematic examination to date of Australian attitudes in this area would appear to be by Wilson & Brown (1973). The authors made an Australia-wide survey of what type of sentence people thought appropriate for various crimes. A great deficiency in their work, however, was the absence of any attempt to compare the data they gathered with actual sentencing practice. They ascertained what the people wanted but failed to find out whether that was what the people were getting. Additionally, they presented their findings in a categorised form which makes comparison with official statistics very difficult.

The reason why Wilson & Brown failed to compare sentencing practice with community standards is not far to seek. The publicly available statistics on sentences awarded by the courts are very sketchy. They are presented only in a highly summarised form which obviously ignores the complexity of the individual cases which form the raw material of sentencing practice.

In spite of the acknowledged difficulties involved, however, it still seems worthwhile to attempt some examination of what is, after all, a question of obviously great public interest. It is very common for some outcry to be raised in the newspapers about sentences that appear to be too lenient by public standards. What, however, are public standards? Are they what newspaper proprietors and journalists decree them to be or can we provide more objective information? Is sentencing practice in fact too lenient by public standards? Some answer to this is presented below.


It was desired in the present study to examine public punitiveness both in the abstract and in relation to particular crimes. It was also desired to determine whether anxiety or neurosis was a determining factor in punitiveness. House & Wolf (1978) have shown that mistrust of others increases as the crime rate in a particular area increases. This suggests that fear is a potent factor in public attitudes. What effect might we therefore expect from those who are chronically fearful or anxious? Might not their fear lead them into advocacy of measures designed to eliminate sources of fear from public circulation for as long as possible? In other words, is punitiveness at least in part, a manifestation of personal neurosis? Other influences on punitiveness might be demographic factors such as age, sex, education and occupation. The extent to which one had personal acquaintances with criminals could also have some influence. The present study was designed to give some examination to all these possibilities.

A questionnaire was prepared which began with the 20 items of the Spielberger et al. (1970) State-anxiety scale, continued with twelve items of a specially written attitude scale, designed to measure punitiveness in general terms, and ended with 21 items which were brief descriptions of particular crimes. After the attitude questions, four basic demographic questions were asked, plus a single question reading: 'How many people, to your knowledge, have you ever met to talk to who have done a spell in jail?'

The Spielberger anxiety scale appears to be the most widely-used measure of chronic anxiety at the moment. The Eysenck (1967) neuroticism scale is also widely used, but unlike the Spielberger scale it lacks any balancing (non-neurotic) items. As such, its scores could be unduly affected by acquiescent response set. The punitiveness scale was designed as an example of the traditional type of attitude measurement -- a Likert scale. Likert scales have widely acknowledged strengths and are well adapted to enable an examination of what people think in the abstract about a particular type of issue. The third group of questions represented an attempt to describe particular crimes realistically and vividly without at the same time degenerating into the ungeneralizably particular. They were preceded by the instruction: 'We would like to list certain crimes and then hear how many years of prison you think each deserves. If you think that the crime should be punished by some sentence other than prison, just say "zero-years" of prison'.

The questionnaire was administered by trained and supervised student interviewers to a random cluster sample of the Sydney metropolitan area. A total of 113 people were interviewed. The distribution of demographic characteristics observed in the sample corresponded closely to those of previous Sydney samples. The mean age was 36 years, there was a slight preponderance of females, mean education fell between the School Certificate and the Higher School Certificate and 62% worked in non-manual occupations. The sample size may seem small by public opinion poll standards but this is simply because accuracy of estimates down to one tenth of one per cent were not considered essential on the present occasion. For detecting significant mean differences or significant correlations, the given sample size was ample. For instance, a correlation explaining as little as 3.5% of the common variance would be shown as significant at the .05 level with the given N. The sampling methodology was the same as that used by all but one of the Australian public opinion polls (Ray, 1975).


With all three groups of items scored as summative scales, the reliabilities observed (alpha) were .84 for the Spielberger scale, .83 for the Punitiveness scale and .77 for the Length-of-Sentence scale. The Length-of-Sentence scale correlated a significant .302 with the Punitiveness scale -- thus providing some validation for the Punitiveness scale. The Anxiety scale, however, failed to correlate significantly with either the Punitiveness or the Length-of-Sentence scale. This indicates that punitiveness in neither the general, nor the particular sense is an outcome of personal neurosis. There were also no significant correlations between demographic variables and either type of punitiveness. Knowledge of criminals was unrelated to both senses of punitiveness and also to anxiety.

In order to compare the length of sentence desired with the length of sentence actually prevalent in the courts, an item-by-item approach is necessary. Below then are listed the survey items in their order of occurrence together with such comparisons with official statistics as can be made.

1. An Aboriginal with little education seizes a six-year-old white girl and rapes her, so injuring her that she bleeds to death.

This item was explicitly intended as a concise description of a particular notorious crime -- the case of Rupert Max Stuart in South Australia. The description was in fact recognised by many of the respondents and was generally remarked upon with recognisable anger. The mean length of sentence awarded by the respondents was 48 years. Stuart in fact served fifteen years in prison. As this was an exceptional crime, no comparison with official statistics seemed possible. For further details see Graves (1973).

2. A prominent banker defrauds a large charity of a million dollars which he spends on gambling.

This item was also intended as a description of a particular crime -- that of Rural Bank head Peter Huxley. The mean sentence awarded by the respondents was seventeen years. Huxley in fact served eight years in prison. As this was again an exceptional crime, further comparisons seemed superrogatory. It may be worth noting that Huxley was originally sentenced to 20 years with a twelve-year non-parole period. He also took part in a work-release scheme before his full release (McCarthy, 1978).

3. A pack of young men who work as labourers pick up a girl in a hotel and offer her a lift home. They then take to an isolated spot and rape her. They threaten to kill her if she tells the police but then let her go.

This and the following were intended as descriptions of more typical crimes. There should therefore be some point in comparing the distribution of sentences revealed in Statistics of Higher Criminal Courts for NSW published by the Australian Bureau of Statistics. In 1977 (the latest year available to me) there were 43 persons convicted of rape in NSW (Table 10). Their sentences are given in category form only but, by taking the midpoint of the category as the category mean, an overall mean can be calculated. 'Ten years or more' was taken as an average of 12.5 years. The rough mean that emerges then is 8.6 years. This of course is only the sentence. Allowing a reduction of roughly 1/3 for parole, we arrive at an approximate average time served of 5.7 years. The average sentence awarded by the respondents in the survey was 30 years. 76% of the survey respondents awarded sentences of ten years or more compared with 30% in the actual court statistics.

4. A professional burglar breaks into the home of a working couple and steals their TV for resale in a pub.

This offence ('Break, enter and steal') is covered by both lower and higher courts. In Court Statistics 1979 published by the NSW Bureau of Crime Statistics & Research we have the lower court statistics. If those who served less than one year are taken as a mean of six months and those who served one to two years are taken as a mean of 1.5 years, we arrive at an overall average sentence of approximately nine months for a total of 184 offenders. Offenders not given a prison sentence are omitted from these calculations. If these 587 offenders are included (as they should be for comparability with the survey statistics) the mean sentence drops to two months. Statistics of Higher Criminal Courts reveals 132 offenders under 'Break, enter and steal' who were not imprisoned and 277 who were. Again using the category midpoint from Table 10 as the category mean and assuming a mean of 12.5 years for the (only two) offenders given 'Ten years or more', a mean length of sentence of 2.2 years for 409 offenders is arrived at. If we calculate a weighted average of higher and lower court statistics, the mean sentence is just less than one year. The sentence awarded by the survey respondents was seven years.

5. A carpenter working for a large oil company steals a quantity of timber which he uses to do up his house.

This offence would seem to correspond most closely to the lower court offence of larceny. From Table 2.3 of Court Statistics 1979 we find that of 9738 cases where the offence was proven, 728 received sentences up to one year and 19 of one to two years. This represents an average jail sentence of two weeks. The sentence imposed by the survey respondents was 2.5 years.

6. A shop assistant working for a large department store regularly takes home goods from the store until she has far more than she ever could use. When caught, she says she couldn't help it.

This offence is particularly hard to trace in official statistics. 'Shoplifting' is nowhere tabulated as a separate offence. Perhaps the most nearly comparable category is the lower court offence 'Unlawful possession of property'. There were 806 cases where this offence was found proven but was not punished by imprisonment. 120 were imprisoned for up to a year and 5 for one to two years. This represents an approximate mean sentence of three weeks. The mean sentence awarded by the survey respondents was 1.1 years.

7. A lady from a respectable family with two grown children takes a dress from a department store and tries to get it out without paying for it.

While the preceding offence (6 above) is best described as 'pilfering', this offence (7) is shoplifting proper. The same difficulties in detecting how the offence is treated by the courts therefore arise. The mean sentence awarded by the survey respondents was 7.5 months. The lower court offence of larceny is perhaps a near equivalent. If so, an average jail sentence of two weeks (see 5) was awarded by the courts.

8. The manager of a factory which produces poisonous wastes carelessly lets some of the waste escape into a river which runs nearby.

The only category of environmental crime is found in the lower court statistics. Of the 189 proven offences, none were punished by imprisonment. The survey respondents awarded an average sentence of eight years.

9. A group of young louts beat up an old man they come upon in a bus shelter 'just for kicks'.

In the category of major assault in the higher court statistics, we find 73 offenders not imprisoned and 70 imprisoned. Calculating mean sentence as before yields a figure of two years. This compares with twelve years awarded by the survey respondents.

10. A drunken driver knocks down a policeman and kills him.

This offence was included as a description of an incident that occurred during the author's boyhood in Cairns, North Queensland. The offender in that case was simply fined 50. Whether the leniency of the sentence was due to the unpopularity of the police or to the notorious reluctance of Cairns juries to convict locals must however remain speculative. In the higher court statistics under driving offences occasioning death, we find 13 not imprisoned and 35 imprisoned. There were also twelve sentences of periodic detention. If we regard those offenders sentenced to less than one year and those sentenced to periodic detention as both serving a notional sentence of six months, the mean sentence was 1.5 years. The survey respondents awarded 16.5 years.

11. A man driving too fast runs over a child and cripples her for life.

Under driving offences occasioning injury in the higher court statistics, we find 33 offenders not imprisoned and 26 imprisoned. Calculating the mean sentence as before gives six months. The survey respondents gave thirteen years.

12. A bus driver driving too fast on a wet road loses control of the bus and it crashes, killing several people.

This also would appear to come under the category of driving offences causing death (see 10 above). The mean sentence awarded by the courts was hence 1.5 years. The survey respondents awarded 6.7 years.

13. A man who earns quite a high income tells lies so that he will not have to pay any income tax.

Taxation statistics appear to be completely confidential. The survey respondents awarded 3.3 years. From what reaches the newspapers, it would appear that imprisonment for taxation offences is very rare in Australia.

14. A clever businessman takes advantage of loopholes in the income tax Act so that he quite legally does not have to pay any tax.

One would think that something that is 'quite legal' could not possibly be punished by imprisonment but the survey respondents awarded an average of one year. Since the Income Tax Assessment Act also on some occasions lumps together evasion (illegal) and avoidance (legal), this sentencing preference may not be as bizarre as it at first appears.

15. A young man with no regular employment holds up a young couple in a park at night and robs them of their money.

This offence, sometimes called 'mugging', would appear to come under the category of 'robbery with minor assault' in the higher court statistics. Fifteen offenders were not imprisoned and eighteen were. The comparatively small number of such offenders makes a rather interesting comparison with the American situation. The mean sentence was 2.2 years. The survey respondents awarded 2.5 years.

16. A secretary persistently robs the petty cash of a firm she works for in order to buy herself nicer clothes.

This offence would appear to be covered by the lower court offence of larceny but could also come under the higher court offence of 'fraud n.e.c.' or 'stealing n.e.c.' As 'n.e.c.' means 'not elsewhere covered', there would obviously be many offences under this category of many degrees of seriousness and comparisons of sentencing practice in such broad categories would tell us little. The lower court offence of larceny would therefore seem to give the most reliable guidelines. The average jail sentence of two weeks for larceny (see 5 above) compares with 1.4 years awarded by the survey respondents. It must be observed, however, that official statistics are particularly uninformative about this, presumably rather common, offence.

17. A young man regularly uses cannabis (pot, marijuana) at parties and on other social occasions.

This would appear to come under 'use or possession of drugs' in the higher court statistics. Reflecting the changing social mores, only thirteen persons were convicted of this offence but eleven were imprisoned. The high rate of imprisonment may suggest that only fairly persistent users of the drug are now brought to court or that it is possession of drugs other than cannabis which is showing up in the statistics of prosecution under this category. It certainly seems to be the case that it is only trafficing in cannabis that is now prosecuted. As the official statistics do not tabulate cannabis convictions separately, however, it is not easy to confirm this. It certainly seems unlikely that the social user (described as 17 above) would ever be prosecuted nowadays. The survey respondents however awarded 2.9 years.

18. A young man becomes addicted to the use of heroin.

The higher court offence of 'use or possession of drugs' should be much more relevant here. The mean sentence in this category was 4.8 years. The survey respondents awarded 4.5 years.

19. A young woman engages in prostitution because it is easy money.

Recent changes in the legislation mean that prostitution is no longer an offence in NSW. The survey respondents however awarded an average jail sentence of nine months.

20. A middle aged man runs an illegal gambling house.

This would appear to come under 'betting and gaming' in the lower court statistics. There were 2682 court actions in this category, however. This suggests that the great majority in this category were the gamblers themselves rather than the organisers. The most relevant information to the present case, therefore, would appear to be the fact that the highest sentence awarded in the betting and gaming category was six months. The survey respondents, by contrast, awarded 1.3 years.

21. A qualified doctor runs a 'no questions asked' abortion clinic which charges $100 per operation.

Since the 'Heatherbrae' case, abortion in properly conducted clinics is no longer illegal in NSW. The survey respondents, however, awarded a mean sentence of four years.


There is very clearly a stark gap between actual sentencing practice and what the public see as appropriate sentencing practice. Difficult though the official statistics are to dissect, the gap between the judiciary and the public is so gross that nothing can disguise it. There would appear to be no possible set of reasonable assumptions which would suffice to close the gap. People in general want far more severe sentences than our courts in fact award.

The obvious inference from this would appear to be that the judiciary (and presumably the class that they represent) are leading public opinion rather than vice versa. Whether or not we have political democracy, there is certainly no democracy in the courtroom. The popular will does not prevail there. Whether this is a good or bad thing must be left to the reader to decide. Those who advocate razing the jails, however, should probably feel rather cast down by the findings. If the judiciary is already so much more lenient than public opinion, the room for further progress in that direction must be very limited. If the gap were to grow any wider, there must surely be serious political consequences.

A matter of considerable interest is to ask how the discrepancy between the judiciary and the public arose. Some obvious considerations are that the sentences awarded by the public could obviously not be as well considered as the sentence of a judge or magistrate who had many hours of evidence before him. Because of the number of recidivists who come before them, the judiciary probably also have a more vivid impression of the ineffectiveness of imprisonment as a deterrent. Thirdly, the judiciary may be better aware of the high cost to the community of imprisonment and the limited prison facilities available. A fourth and more serious possibility is that judges become hardened by the parade of crime that passes before them and lose all sense of how serious crime can be to the victims or potential victims of it.

The findings with the length of sentence data were also borne out by the findings with the punitiveness scale. The overall mean on the scale was 39.48 (SD 7.78) --which is on the punitive side of the scale midpoint (36). This meant that people tended to agree with statements such. as 'Violent crime should be harshly punished' and 'The life sentence for serious crime should mean life' and tended to reject statements like 'All prisons should be pulled down' and 'Prisons should be more like hospitals or schools than they are today'.

Perhaps a matter of considerable interest was the fact that length of sentence preferred and punitiveness were independent of the demographic and personality variables also examined. To be punitive does not mean that you are neurotic, working class, ill educated, old or male. This runs somewhat counter to the findings with the California F scale (Adorno et al. 1950; Brown, 1965). The F scale was supposed to be heavily loaded with punitiveness and was also found to attract higher scores from working class and poorly educated people. As there is however a great deal of doubt over whether the F scale is valid as a measure of what it purports to measure (Ray, 1976; Ray & Lovejoy, 1983) this divergence of findings was not unexpected. The present data, then, consistently counterindicate any theory of punitiveness being psychologically or socially pathological. It is in fact the community norm.

Those who dislike the conclusions of the present study will be pleased to find that there is much in the methodology of the present research that they can query and criticise. It is presumably because of this great difficulty in doing methodologically rigorous research in the area that so little has so far been done. To criticise and to do better are however very different things and only further research can tell whether any of the possible criticisms of the present work have any appreciable impact on the conclusions to be drawn.


Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J. & Sanford, R. N. The authoritarian personality New York, Harper 1950.

Brown, R. Social psychology New York, Free Press 1965.

Eysenck, H. J. The biological basis of personality Springfield 111., Thomas 1967.

Graves, A., Q: What new light can this man throw on the Stuart case? A: None. On Dit 1973, 41 (12) p. 7.

House, J. S. & Wolf, S. 'Effects of urban residence on interpersonal trust and helping behaviour' Journal of Personality and Social Psychol 1978, 36 pp. 1029-1043.

McCarthy, P. 'Huxley: Mystery of the missing 1.8 million' National Times 17-23 Dec. 1978.

Ray, J.J. (1975) Public opinion polls and attitude measurement. Current Affairs Bulletin 52, 24-30.

Ray, J.J. (1976) Do authoritarians hold authoritarian attitudes? Human Relations, 29, 307-325.

Ray, J.J. & Lovejoy, F.H. (1983). The behavioral validity of some recent measures of authoritarianism. Journal of Social Psychology, 120, 91-99.

Spielberger, C. D., Gorsuch, R. L. & Lushene, R. E. Manual for the state-trait anxiety inventory. Palo Alto, Consulting Psychologist's Press 1970.

Wilson, P. R. & Brown, J. W. Crime and the community Brisbane, Univ. of Queensland Press 1973.


The items of the Punitiveness scale.

1 Hanging should be brought back as a penalty for serious crime.
2 Violent crime should be harshly punished.
3 All prisons should be pulled down.
4 A lynching party may be the best way to deal with rapists.
5 Punishment does not deter crime.
6 The prison sentences today are not long enough.
7 The life sentence for serious crime should mean life.
8 More criminals should be made to do hard labour when they are in prison.
9 Prisons should be more like hospitals or schools than they are today.
10 What criminals need is guidance rather than punishment.
11 Prisons should be more humane than they are today.
12 Flogging should have no place in a modern prison system.


See also:

Ray, J.J. (1985) The punitive personality. Journal of Social Psychology 125, 329-334.

Replication is one of the cornerstones of science. A new research result will normally require replication by later researchers before the truth and accuracy of the observation concerned is generally accepted. If a result is to be replicated, however, careful specification of the original research procedure is important.

In questionnaire research it has been my observation that the results are fairly robust as to questionnaire format. It is the content of the question that matters rather than how the question is presented (But see here and here). It is nonetheless obviously desirable for an attempted replication to follow the original procedure as closely as possible so I have given here samples of how I presented my questionnaires in most of the research I did. On all occasions, respondents were asked to circle a number to indicate their response.

Go to Index page for this site

Go to John Ray's "Tongue Tied" blog (Backup here or here)
Go to John Ray's "Dissecting Leftism" blog (Backup here or here)
Go to John Ray's "Australian Politics" blog (Backup here or here)
Go to John Ray's "Gun Watch" blog (Backup here or here)
Go to John Ray's "Education Watch" blog (Backup here or here)
Go to John Ray's "Socialized Medicine" blog (Backup here or here)
Go to John Ray's "Political Correctness Watch" blog (Backup here or here)
Go to John Ray's "Greenie Watch" blog (Backup here or here)
Go to John Ray's "Food & Health Skeptic" blog (Backup here)
Go to John Ray's "Leftists as Elitists" blog (Not now regularly updated -- Backup here)
Go to John Ray's "Marx & Engels in their own words" blog (Not now regularly updated -- Backup here)
Go to John Ray's "A scripture blog" (Not now regularly updated -- Backup here)
Go to John Ray's recipe blog (Not now regularly updated -- Backup here or here)

Go to John Ray's Main academic menu
Go to Menu of recent writings
Go to John Ray's basic home page
Go to John Ray's pictorial Home Page (Backup here)
Go to Selected pictures from John Ray's blogs (Backup here)
Go to Another picture page (Best with broadband)