Political powerlessness and sociodemographic status in Canada*

PETER R. S I N C L A I R 1 University of Guelph

Fonde sur une analyse secondaire de I’enqu2te sur I’election nationale (1974), cet article dkmontre I’association limitee entre I’experience de I’impuissance et la position socio-demographique. I1 offre une explication qui ne desavoue pas l’importance de la situation sociale, mais qui souligne comment le systeme social fonctionne de faGon a produire des alienations differentes en conformite avec I’experience sociale de I’individu. Enfin, il discute l’importance, pour le systeme politique canadien, de la grande proportion des citoyens alienes, e t particulierement la direction ideologique qu’un tel groupe prendrait s’il etait mobilise.

Based on a secondary analysis of the national election survey (1974), this paper shows the limited association between the experience of powerlessness and sociodemographic position. An explana- tion is offered which does not deny the importance of social position, but stresses how the social system functions to produce different kinds of alienation according to the individual’s social experience. Finally, the significance for the Canadian political system of its large proportion of alienated citizens is discussed with particular concern for the ideological direction of such a group if it were mobilized

In these recent years of political crisis in Canada, it is perhaps surprising that empirical research on the extent of political alienation has been minimal compared with the concentration on participation and voting studies. Do Cana- dians trust their leaders and support their politi- cal system? If not, what are the consequences? How widespread is the feeling of powerlessness to control or influence political outcomes? Has the system failed to represent the interests of the people and are there alternatives on the horizon? Surely such questions need to be ad- dressed for an adequate understanding of con-

temporary Canada. In this paper my aim is only to consider part of the problem area. Utilizing the 1974 national election survey,’ I shall analyse the distribution of political powerless- ness or inefficacy2 as it is experienced by con- temporary Canadians, try to account for the observed variations, and assess their sig- nificance for the political system.

THEORETICAL BACKGROUND

The concept of political powerlessness is drawn from the broader theory of alienation. With a

*I am most grateful to Neil MacKinnon for assistance with methodology and to Eleanora Cebotarev, Ken Westhues, Ken Menzies, and the CRSA readers for their comments on an earlier draft. The defects that remain are entirely my responsibility.

1 The 1974 national election study was made available by the Institute of Behavioural Research, York University. The data were originally collected by Harold Clarke, Lawrence LeDuc, Jane Jensen, and Jon Pammett. Neither the Institute nor the original collectors of the data bear any responsibility for the analyses or interpretation presented here. 2 Throughout this paper I shall treat inefficacy and powerlessness as synonyms.

Rev. canad. SOC. & Anth./Canad. Rev. SOC. & Anth. 16(2) 1979

126 / Peter R. Sinclair

long history in medieval theology, enlighten- ment and Hegelian philosophy, Marxism, exis- tentialism, and empirical sociology (Schacht, 1970; Meszaros, 1970; Schroyer, 1973:44-loo), alienation is, however, a category to which n o universal core meaning can be ascribed other than that it always seems to involve some kind of separation (Schacht, 1970:241-42). The liter- ature which discusses the alienation of the indi- vidual from the state or political system is n o less clouded in ambiguity when viewed as a whole, although particular authors may clarify their use of the term and produce valuable analyses. One important division among au- thors is over the use of alienation as a subjective or objective concept. Is alienation to be under- stood as a quality of social relationships in- ferred by the investigator, or is it “real” only when experienced by individuals?

Within the Marxian framework, political alienation may be seen in the former sense, although this is not incompatible with a consid- eration of subjective experience. In his early writings. Marx discusses the relationships be- tween individuals, state, and civil society. Al- though formally distinct, state and civil society are intimately linked. Thus, “from a political point of view the sfale and the organization of society are not two different things. The state is the organization of society” (Marx, 1975a [18441:411). Given that society in question is a class society, it follows that the state “is no- thing more than the form of organization which the bourgeois necessarily adopt both for inter- nal and external purposes, for the mutual guarantee of their property and interests” (Marx and Engels, 1970 [18461:80). The claim that the state’s activities are the pursuit of citi- zens’ common interests is, of course, rejected by Marx as an illusion. Just as the workers are alienated from their products in the productive process, they are alienated as citizens from the institutions which purport to represent their common interests:

In the previous substitutes for the com- munity, in the State, etc., personal freedom has existed only for the individuals who developed within the relationships of the ruling class, and only insofar as they were individuals of this class. The illusory com- munity, in which individuals have up till now combined, always took on an inde- pendent existence in relation to them, and was at the same time, since it was the combination of one class over against

another, not only a completely illusory community, but a new fetter as well (Marx and Engels, 1970 [18461:83).

The state lies external to individuals, indepen- dent of them, and ruling over them. Political emancipation must then involve the overcom- ing of this alienation as part of human emanci- pation generally. “Only when real, individual man resumes the abstract citizen into himself. . . only when man has recognized and organized his forces propres as social forces SO that social force is no longer separated from him in the form of political force, only then will human emancipation be completed” (Marx, 1975b [ I 8431:234).

The considerable volume of research on polit- ical alienation which has been published since 1950 usually acknowledges an intellectual debt to Marx, although the focus of this work has been on the extent to which peoplefed alien- ated. This has not been accompanied for the most part by a critique of the structure of soci- ety which generates “objective” alienation and camouflages its existence from some part of the population. My purpose in this essay will be to begin such a critical analysis of the experience of political powerlessness in contemporary Canada.

A necessary preliminary task, however, is to establish a more exact concept. With regard to measures of political alienation, Citrin e t al. (1975:3) state that “taken as a group, the politi- cal alienation scales found in the research litera- ture refer to every kind of negative feeling about political institutions and officials.” And Mil- brath and Goel (1977:62) hardly clarify its mean- ing when they write that “political alienation may be defined as deep-seated and relatively enduring feelings of estrangement, rejection, negativism, and unhappiness with the political system and its salient parts.” One response to the problem has been to specify particular “modes” or “dimensions” of political aliena- tion (Keniston, 1965; Aberbach, 1969; Finifter, 1970; Gamson, 1968). Ada Finifter’s work is of particular importance for her empirical spec- ification of these “dimensions.” In her concep- tual framework, four “modes” of political alienation are distinguished. “Political powerlessness” refers t o “an individual’s feel- ings that he cannot affect the actions of the government . . . Political decisions, which de- termine to a great extent the conditions under which an individual lives, may appear to be

Political powerlessness and sociodemographic status / 127

huppenirig to individuals who feel powerless, independent of or in spite of their own judge- ment or wishes” (Finifter, 1970:390). “Political meaninglessness” refers to a condition in which decisions are thought to be random or unpre- dictable, while “political normlessness” is said to exist when people perceive “that the norms or rules intended to govern political relations have broken down, and that departures from prescribed behaviour are common” (Finifter, 1970:390-91). This is very like the concept of distrust or cynicism used by such authors as Aberbach (1969), Aberbach and Walker (1970), Easton (1973, and Miller (1974). Finally, “political isolation” is defined as “a rejection of political norms and goals that are widely held and shared by other members of a society” (Finifter: p. 391). When Almond and Verba’s “civic cultures” data were reanalysed by Finif- ter using factor analytic techniques, two “modes,” labelled “political powerlessness” and “political normlessness,” were revealed.

Such efforts have not gone without criticism. For example, in discussing the general concept of alienation, Schacht (1970: 155-56) argues against this strategy of establishing separate “dimensions,” which are conceptually distinct, yet labelled as part of a single phenomenon. Fischer (1976) too is concerned to point out that his reformulation of alienation involves spec- ifications but not dimensions of alienation. He includes powerlessness as part of his concep- tualization, but assigns cynicism or distrust to a separate category. My own preference would be to limit the use of political alienation as a social psychological concept to the experience of political powerlessness. This would provide a clear link with the Marxist tradition, as well as consistency with most recent definitions of political alienation. But political alienation has been used to cover too much conceptual terri- tory. Therefore, in the interests of clarity, the more ungainly term “political powerlessness” will be used here following Finifter’s definition.

Early research on the social structural corre- lates of political powerlessness or inefficacy was conducted at the Michigan survey research centre from where Campbell et al. (1954; 1960) reported cross-tabulations with sociodemo- graphic categories such as class, region, age,

sex, etc. This tradition is still alive in the recent report from Gilmour and Lamb (1975) who show that, in the late sixties and early seventies, feelings of powerlessness and distrust in the United States were generally more common among relatively low status, peripheral, and objectively deprived groups. They did, how- ever, note the presence of alienation among the more privileged sections of the population. Others have reported widespread alienation among conservatives in the south and west (Phillips, 1970) or among middle-status Ameri- cans threatened by rapid social change (Scam- mon and Wattenberg, 1970). It is not surprising, then, that Converse (1972), reviewing the Michigan studies, should feel able to claim that alienation increased equally in all main sections of the population. House and Mason (1975) have significantly developed Converse’s notes and point to the weakness of sociodemographic categories to explain alienation.

Standard demographic variables such as race, region, income, sex, education and age have been central to most analyses of the social bases of voting and political at- titudes. But, the continued analytic useful- ness of these traditional demographic clas- sifications for understanding aspects of re- cent political behaviour should clearly not be assumed (p. 145).

In Canada, studies of powerlessness based on the 1%5 and 1968 national election surveys have stressed class, education, and especially region as major independent variables (Simeon and Elkins, 1974; Schwartz, 1974). In this paper I shall pay brief attention to temporal changes and concentrate on the sociodemographic cor- relates of powerlessness in 1974 to show ( 1 ) that low status and peripheral groups are more alienated; (2) that region3 of residence is more critical than the cultural variables of religious affiliations or ethnic identity; and (3) that pow- erlessness is nevertheless widespread across sociodemographic categories. Finally, some implications of these results for the operation of the Canadian political system will be set out together with some suggestions as to the reasons for the failure of sociodemographic status variables to serve as good predictors of powerlessness.

3 As Simeon and Elkins (1974:399n) suggest, region per se explains nothing. We should treat region as an intervening variable and search for the patterns of social development which may account for regional variations.

128 I Peter R . Sinclair

METHODS

My analysis is a secondary investigation of the 1974 national election study, which was a post- election, nation-wide survey of 2562 respon- dents. The sample design called for a multi- stage, stratified, cluster sampling with overrep- resentation of the smaller provinces. (For afull description, see Leduc et al., 1974.) In order to produce a valid national sample, a weighting procedure was followed which reduced the ef- fective size to 2,445 persons. When analysing individual provinces or regions, the unweighted subsample is used with one exception. For the Atlantic region it is necessary to adjust for the excess of PEI respondents. Unfortunately, the length of the survey instrument dictated that certain questions could be asked only of a ran- domly selected, half sample, which, in the case of the items on powerlessness, amounted to 1,262 respondents.

In the survey, five items were included as a measure of political efficacy. Of these, three had been used since 1952 in the Michigan presi- dential election studies and in the two Canadian

investigations of 1965 and 1968. All five items (listed in Table 1) have face validity as measures of perceived powerlessness, although the fifth has not been included by Leduc (1976) in his efficacy scale.

In order to check that the variables did indeed reflect a particular theoretical construct, a prin- cipal components analysis was undertaken. This produced the anticipated single factor so- lution. However, given that the coefficient of reliability was only moderately high (.68), it was decided to use factor scores to measure political powerlessness. In this way, each of the five items contributes to the score on powerlessness according to its contribution to the principal component4 and is preferable to one based on the simple averaging or summing of scores on items. The factor score variable is in standard format with negative scores representing grea- ter experience of powerlessness.

The independent variables in the analysis include the sociodemographic categories which earlier students have considered as causes of the experience of powerlessness, inefficacy, or alienation. They are measured as follows:

TABLE 1

Level of Political Powerlessness in Canada, 1965-74, and in the Regions, 1974

Percentage who azree that:

Nut. * Nut. 7 ’Nut. Atlantic Quebec Ontario Prairies BC 1965 1968 1974 1974 1974 1974 1974 1974

Generally, those elected to parliament soon lose touch with the people. __ 56 65 64 72 63 62 64

I don’t think the government cares much what people like me think. 46 42 58 61 67 54 59 42

Sometimes, politics and government seem so complicated that a person like me can’t really understand what’s going on. 69 69 66 72 66 63 77 56

People like me don’t have any say about what the government does. 49 47 55 54 69 47 58 38

So many other people vote in federal elections that it doesn’t matter very much whether I vote or not. __ - 14 20 22 10 14 6

Number of resoondents 2727 2767 1203 252 341 344 198 127

*Source of 1%5 data: Schwartz, 1974: 224. ?Source of 1%8 data: Simeon and Elkins, 1974: 407,411. For confidence intervals, see Appendix 1 .

4 Factor coefficients for the five items are: MPs lose touch Government doesn’t care Politics too complicated People have no say So many vote

0.30 0.35 0.25 0.35 0.21

Political powerlessness and sociodernographic status / 129

Education: Number of years in school. Occupational status: The four-digit Blishen scale of occupational prestige. Family income: Respondents were asked to place their family income in one of nine categories with low values representing low income. Region: The conventional classification of provinces into Atlantic, Quebec, Ontario, Prairies, and British Columbia is used. In the regression analysis, region is treated as a set of dummy variables with British Columbia as the reference category. (See Nie et al., 1975.) Community size: The survey classified respon- dents according to a combination of size and type of community. Here only the population criterion is used in a seven-point scale in which a score of one indicates residence in a city of at least 500,000 persons and seven indicates rural residence. Ethnic identity: As a rough indicator of ethnic identity, language spoken at home is adopted here. English, French, and other languages are represented as dummy variables with “mixed” language homes as the reference category. Religious identity: Catholics, established Pro- testants (Anglicans, United Church, Pres- byterians, Lutherans), and Jews are dummy variables for religious identity with others as a reference category. Sex: A dummy variable scored one for males. Age: Measured in years.

In order to determine the relative importance of the independent variables as predictors of pow- erlessness, multiple regression analysis was conducted, including in the equation in succes- sion the social status variables, territorial vari- ables (region and community size), the cultural variables (religious and ethnic identity), sex, and age.

FINDINGS

The percentages agreeing with the five items on powerlessness are presented in Table 1 where it is apparent that, with the exception of the vot- ing item, levels of powerlessness were high, and furthermore, had increased in three of four instances between 1968 and 1974. This probably reflects the deteriorating situation of the economy compared with the late sixties. In- deed, inflation was by far the most important issue in the minds of the electorate as fully 56 per cent mentioned it in response to an open-

ended question (Pammett et a]., 1977:102). Also, the next three most popular responses were related issues: wage and price controls, economic issues, pensions, and taxes. While economic questions do not seem to have been a dominant basis for the voters’ party choice (Pammett et al., 1977:107), they may well have contributed to a growing feeling that govern- ment was distant from and could not be influenced by ordinary people.

The fifth item (“So many other people vote . . . ”) calls for special comment. While most Canadians feel that it matters whether they vote or not, this does not mean, if we can believe the other items, that they feel able to exert much influence on the system. Only the extremely alienated voters opted out of the system al- together. When the voting question was phrased differently in 1965, 75 per cent agreed that “voting is the only way people like me can have any say about how the government runs things” (Schwartz, 1974:224). It is likely that most of those who agreed with this statement in 1965 would also have accepted that it did matter whether they voted or not. People feel they should make whatever small input they can. Unfortunately, since the older item was not asked in 1974, this inference cannot be checked.

Table 1 also presents the level of powerless- ness by region. Although in some cases the small sample size should make us cautious about interpreting percentage differences (see Appendix l), the gap between the extreme re- gions is worth noting. In four of the five items, Quebec demonstrates the highest incidence of political powerlessness, British Columbia the least; and the difference reaches as much as thirty-one points for: “People like me don’t have any say about what the government does.” Compared with the sixties, it appears that Quebec has taken over from Atlantic Canada as the most disaffected region, while British Columbia retains its former position. The findings on Quebec are consistent with the rapid rise of Quebec nationalism and the popu- larity of the Parti quebecois. It would be in- teresting to know whether feelings of power- lessness declined during the early years of that party’s rule in Quebec.

The hypothesis that low status and peripheral regions encourage feelings of political power- lessness among their residents is corroborated by the multiple regression analysis reported in Table 2. With only social status variables in- cluded (i), the regression equation accounts for

130 / Peter R. Sinclair

TABLE 2

Multiple Regression Analysis of Political Powerlessness, 1974

Standardized regression coefficients lndependeni variables

Education Occupational status Family income Atlantic Quebec Ontario Prairies Community size English ethnicity French ethnicity Other ethnicity Catholic Est. Protestant Jewish Sex Age

(i) (ii)

.27* .25*

.06 .06

. 1 1 * .08 - .04 --.26* -. .09

10*

(iii) (iv) ( v )

.25* .25* .25*

.08 .08 .08

.09* .09* .09* - .06 - .06 - .07 -.26* -.24* -.25* - . l o -.I0 - . 1 1 -.12* -.12* -.13*

.06 .06 .07 -.03 -.01 - .05 - .09 - .02 -.03

.08 C.01 - .02

(vi)

.25*

.08

.09*

~

- .07 -.25* - . 1 1 - .13*

.07 - .02 - .09 -.03

.08 <.01 - .02 -.01 -.01

R2 .134 .169 ,172 .172 .177 .177

*Coefficient at least twice the standard error.

13.4 per cent of the variance in powerlessness. Comparing betas, education has the strongest effect (.27), followed by family income. It is noteworthy that occupational status has a neg- ligible and statistically insignificant effect, perhaps due to its dependence on education and income which are measured separately. The zero-order correlation of political powerless- ness and occupational status, however, is .26, one of the stronger relationships in the data.

When region is added to the equation, the set of dummy variables together raise R 2 to .169 (3.5 per cent). As expected from percentage scores, Quebec is the main contributor. A further small but statistically insignificant in- crement is achieved by the inclusion of com- munity size (iii). The addition of further vari- ables (ethnic identity, religious affiliation, sex, and age) raises the explained variance by only 0.005 per cent and no coefficients are significant (vi). The zero-order correlation of powerless- ness and French ethnic identity is -.21, but this is due almost exclusively to the concentration of French in Quebec and is rather small in any case. Contrary to popular belief, French ethnic- ity is not closely linked with powerlessness, or no more so than can be explained by the rela- tively low social standing of French Canadians.

Ethnicity, like region, may well be important, however, as a focal point for the underlying bases of discontent.

Separate analyses, reported in Table 3, were conducted to see whether national level data masked the salience of certain variables in par- ticular regions. Few coefficients are statistically significant, but certain differences are suggested. Occupational status (unimportant at the national level) is the strongest predictor in Quebec and Atlantic Canada. In the latter re- gion, people in smaller settlements are also more inclined to report an experience of politi- cal powerlessness. In Quebec, education is not significant, while in Ontario it is the only sig- nificant independent variable and is very strong (beta = S1) . In general, the trend in the data is for education to be much more important in Ontario than in the rest of Canada with regard to predicting powerlessness. Conversely, occupa- tional status and family income are somewhat more important outside Ontario. Put another way, in Ontario, low-status people feel power- less only if they have little education; elsewhere, the same group feel powerless even if highly educated. Why this should be so is puzzling.6 It is possible that the federal government’s pro- grams of regional development and the prom-

5 The problem of multicollinearity makes the coefficients of the variables unstable. 6 Since, for some items, Schwartz (1974) found curvilinearity in the 1%5 data, scattergrams for education with powerlessness were inspected for each region. Curvilinear relations were not apparent.

TA

BL

E 3

Mul

tiple

Reg

ress

ion

Ana

lysi

s of

Pol

itica

l Pow

erle

ssne

ss b

y R

egio

n, 1

974

Atla

ntic

Q

uebe

c O

ntar

io

Prai

ries

B

C

Inde

pend

ent

vari

able

s b

Bet

a r

b B

eta

r b

Bet

a r

b B

eta

r b

Bet

a r

Edu

catio

n -.0

3 -.

09

.I1

.02

.06

.32

.14*

.5

1 .3

7 .0

8 .2

3 .3

7 .0

7 .2

0 .2

8 O

ccup

atio

nal s

tatu

s .0

2*

.29

.27

.02*

.3

0 .3

5 -.

01

-.

17

.I3

.01

.19

.38

.01

.11

.I8

Fam

ily in

com

e .0

4 .0

8 .2

3 .0

3 .0

5 .2

3 .0

2 .04

.I8

.07

.17

.27

.04

.08

.21

Com

mun

ity si

ze

-.15

* -.2

5 -.3

1 .0

6 .I

2 .0

1 .0

5 .I

0 .0

3 -.04

-.lo

-.2

3 .1

2 .2

2 .2

0 E

nglis

h et

hnic

ity

-.43

-.18

-.03

-.23

-.06

.0

9 -.

19

-.01

.0

5 -.2

1 -.0

1 .0

2 p.

61

-.01

.I5

2 -. c. Fr

ench

eth

nici

ty

-.16

-.06

.l

o

-.46

-.I7

-.

I0

-.66

-.01

-.

05

-.87

-.01

-.0

3 -

-

-

Oth

er et

hnic

ity

-.78

-.21

-.lo

-.7

6 -.0

2 .0

8 -.7

5 -.

I2

-.09

1.12

.I

3 .I

2 -

-

-

E

6 -3

Cat

holic

-.0

9 -.

05

.04

.29

.09

-.01

-.

04

-.02

.01

.38

.19

.06

.44

.I5

.I3

Est

. Pro

test

ant

-.06

-.0

3 .OO

.27

.06

.05

-.30

-.16

-.08

.15

.08

-.01

.1

3 .0

6 .0

1 Je

wis

h -.4

5 -.

03

-.01

-.90

-.07

-.09

-.

01

c.01

-.0

3 -

-

-

-

-

-

I/) P B 2

m

Sex

.27

.16

.07

-.I7

-.0

8 -.

I1

.03

.01

.04

-.I5

-.09

-.11

.2

6 .1

4 .I

7 A

ge

-.01

-.I1

-.

03

-.0

1

-.I1

-.

I7

<.01

.0

7 -.1

3 <.

01

.03

-.I8

C

.01

-.03

-.13

m

Con

stan

t .6

9 -1

.15

-1.1

0 - 1.

72

-1.2

8 R

2

,201

.I

80

,184

.2

54

,181

m

*Coe

ffic

ient

at l

east

twic

e th

e st

anda

rd e

rror

.

a

0 g.

132 /Peter R . Sinclair

ises of successive regional politicians have raised aspirations considerably. When com- bined with beliefs that education brings ad- vancement, the result is that those in Canada’s peripheral regions are particularly frustrated when education does not indeed bring other rewards. Thus, the “normally” dampening ef- fect of education on powerlessness is countered by other circumstances.

DISCUSSION

What is most notable about the data is not the relative importance of social status and terri- tory, but the very modest contribution of the total package of sociodemographic variables towards explaining political powerlessness. This is not unusual, as House and Mason (1975) account for no more than 20 per cent of the variation in political alienation in the United States, while Finifter (1970) explained 50.7 per cent largely by including political participation among the predictors. As Milbrath and Goel (1977:58-59) report, this particular relationship between participation and efficacy is one of the best documented in the literature, although it is not at all clear what the causal connection is. It is plausible that those who feel powerless par- ticipate less because they see it as wasted ef- fort. On the other hand, perhaps the sheer act of participating enhances one’s feeling of efficacy. In any case, the problem of the weak explana- tory power of sociodemographic variables is not affected unless the first account is correct. It would then be possible that their direct relation- ship with participating might indirectly raise their impact on political powerlessness. In Canada, this link would still be weak as Burke et al. (1978:71), using the 1974 data set, report correlations between participation and efficacy o f . 15 at the federal level and .12 at the provin- cial level. The problem must be addressed further.

One possibility, often mentioned and quickly forgotten in such studies, is measurement error. This is especially likely with attitudinal vari- ables like political powerlessness, which are not subject to easy, unambiguous observation. At- titudes are complex phenomena. not easily communicated in the “artificial” and truncated dialogues of the social survey. Responses to questions, posed by a stranger, on matters which may have been given little prior thought become the basis for statistics on the distribu- tion of powerlessness. We are obliged to as-

sume that the questions mean the same thing to each subject, that they are always treated seri- ously and that each agree or disagree has the same meaning. Such problems may affect the results of our statistical manipulations, particu- larly if measurement problems are greater for some sociodemographic statuses than others. My conclusion is not that the survey method is useless, but that Canadian political sociologists should consider more intensive field methods to probe actors’ conceptions of power and poli- tics. Thus, there is a strong case for using the survey as a starting point from which to begin more intensive research. Quantification should be treated as a beginning rather than an end, although I do not deny that such “open-ended” investigation may contribute, in the long run, to more sophisticated and successful quantifica- tion.

Nevertheless, there are several reasons why, even when using a superior methodology, we might expect that sociodemographic categories would be poor discriminators with regard to political powerlessness. Let us consider first the low-status, objectively deprived groups. In- sofar as they are aware of the potential use of the political system to redress their cir- cumstances, and insofar as they are conscious of their relative deprivation, we would expect persons of low socioeconomic status to feel more powerless than others. Our findings suggest that they do to some extent, as our measures of education, family income, and re- gion are significant predictors of powerless- ness. It is probable that some combination of the political socialization and life experience of the low-status, “politically powerless” people leads them to define the system as too elitist in orientation. Many, however, do not report that they feel powerless. They have, to use Marxist language, a false political consciousness, but one which is not irrational. Such persons are likely to have incomplete knowledge of how the political system works to limit redistribution of income and wealth, maintain unequal access to social and educational services, etc. They see the improvements over time and not the dis- tance still to be covered. (It is possible that the more intensive research procedure advocated earlier would lead some to reconsider their posi- tion. In a sense, researchers may create the world they study.)

The other side of the problem is the experi- ence of higher status individuals. It is not sur- prising that those who derive greater benefits

Political powerlessness and sociodemographic status / 133

from the structure of society should feel less powerless. What is surprising at first glance is the extent to which the objective beneficiaries of a political system which maintains their status exhibit “false consciousness.” We may hypothesize that they perceive such things as the disappearance of their income in taxation to support “welfare bums,” but not the regressive nature of the taxation system as a whole nor the stabilizing impact of social service expendi- tures (see Piven and Cloward, 1971). They may recognize, correctly, that their power as indi- viduals is generally very small, but fail to rec- ognize their power as a class or stratum. Perhaps, having been socialized to accept val- ues of individualism and minimal government intervention, they see the actual performance of the political system as excessively populist or collectivist. Therefore, they come to feel dis- tant from it and powerless. The higher status, “powerless” category may also include a small group of ideologically radical individuals, who have become conscious of social inequalities, but who feel unable to change them by working within the existing political system. Both groups feel powerless, but for different reasons and with different implications. More spec- ifically, if the former group can be mobilized, they will represent a conservative and probably authoritarian force, whereas the latter will be more attracted to radical, egalitarian causes. Obviously, these speculations, and I can claim no more for them than that, should not be accepted until tested by a variety of research procedures.

Distinct from the question of causality, is that of the significance to Canadian society of wide- spread powerlessness, however caused. What are the implications for political action? Is the stability of the system threatened by the “pow- erless” or is it maintained by their inactivity? From earlier research, summarized in Milbrath and Goel (1977:69-70), there appears to be no straightforward answer. It appears to be the interaction of powerlessness and distrust which defines the individual’s level of support for the political system. Based on this premise, Simeon and Elkins (1974:409) have produced for Canada the simple typology summarized in Fi- gure 1 . If the analysis so far is accepted, it follows that powerlessness (low efficacy) does not threaten the system, since those who feel this way are either inclined to be deferentials, when they also trust the government, or to withdraw from political life, when they are dis-

FIGURE 1

The Simeon-Elkins Typology of Citizens

TRUST

High

EFFICACY

trustful. It is the combination of high efficacy and distrust that creates the conditions of or- ganized radical protest. Since powerlessness and distrust are positively related in most studies, the proportion of critics and deferen- tials tends to be small. Because the trust items were not included in the 1974 study, the rela- tionships established by Simeon and Elkins cannot be checked exactly, but Kornberg et al. (1976) find a correlation of .21 between their efficacy scale and support for the political re- gime (measured by asking respondents to rate their feeling about the government of Canada on a “thermometer” from 0 to 100).

The situation I have described is less stable than would appear at first glance because the apathetic mass of powerless and cynical citi- zens are a population available for mobilization. When their discontents become articulated by leaders who can make them believe that change is possible, the withdrawn population may be- come a radical, social movement. This has oc- curred in khe past on numerous occasions as disaffected Canadians have been mobilized to form protest movements on behalf of labour rights, prohibition, native peoples, agrarian in- terests, Quebec nationalism, etc. (see Clark et al., 1975). It is interesting to speculate at this juncture why the Atlantic provinces, so de- prived by objective criteriarelative to the rest of Canada, have been remarkably quiescent (with some exceptions such as the Cape Breton min- ers and steelworkers). The discontent has been there, but it has rarely been mobilized, perhaps because that famous export of the region, its people, has left the remainder deprived of their most educated, innovative, and articulate spokesmen.

If the argument so far is accepted, then, in the absence of improved conditions, we can expect that populist attempts to mobilize the discon- tented will meet with some success. One key

134 / Peter R . Sinclair

problem is that these movements can be either authoritarian or democratic, and a previous history of relatively democratic development is no guarantee for the future. Movements such as the Danish Progress Party can erupt into an apparently orderly political scene and threaten established values (Borre, 1974; Nielsen, 1976). This party, standing for the elimination of in- come tax, together with a drastic reduction in public services, collected 16 per cent of the vote in 1973. Supporters were highly alienated and authoritarian. In Canada, the Social Credit re- vival in British Columbia comes closest to reflecting this ideology, and there are elements of it in, for example, the Liberal Party of On-

tario’s appeal in the 1977 election for a return to “basics” and less government spending. The long-run impact of austerity policies favoured by right-wing parties may be an increased level of social conflict in that, paradoxically, they make it easier to mobilize the socialist and populist left. A recent example is the October 1976 strike, organized by the usually passive Canadian Labour Congress, to protest the fed- eral wages and prices policy.8 Far from living at the end of an ideology era, we may well be entering a period of class polarization and conflict, fuelled in part by the mobilization of previously apathetic and “powerless” indi- viduals.

APPENDIX 1 Since standard tables of confidence limits refer to simple random samples and since cluster sampling tends to increase sampling error, fhe following table should be used to assess percentage differences.

_____ Ninety-five Per Cent Confidence Limits for Percentages by Size of Base

For percentages N 5 or 95 10 or 90 20 or 80 30 or 70 40 or 60 50

~ ~ ~~~~

100 5 7 9 11 11 12 200 4 5 7 8 8 8 400 3 4 5 6 6 6 600 2 3 4 5 5 5 800 2 3 4 4 4 4

loo0 2 2 3 4 4 4 1500 2 2 3 3 3 4

REFERENCES

Aberbach, Joel D. 1969 “Alienation and political behaviour.”

American Political Science Review 63 (March) : 86-99

Aberbach, Joel D., and J.L. Walker 1970 “Political trust and racial ideology.”

American Political Science Review 64: 11 99- 12 19

Borre, Ole 1974 “Denmark’s protest election of De-

cember 1973.” Scandinavian Political Studies 9: 197-204

LeDuc

pation in Canada: some methodological

Burke, Mike, Harold D. Clarke, and Lawrence

1978 “Federal and provincial political partici-

and substantive considerations.” Cana- dian Review of Sociology and Anthro- pology 15( 1):61-75

Campbell, Angus et al. 1954 The Voter Decides. Evanston, Ill.: Row

1960 The American Voter. New York: Wiley Citrin, Jack, H. McClosky, J.M. Shanks, and

P.M. Sniderman 1975 “Personal and political sources of politi-

cal alienation.” British Journal of Politi- cal Science 7 (January): 1-32

Clark, Samuel D., J. Paul, and Linda M. Grayson (eds.)

1975 Prophecy and Protest: Social Movements in Twentieth Century Canada. Agin- court: Gage

Converse, Philip

Peterson

7 For elaboration of the concept of populism and its authoritarian and democratic variants in western Canada, see Sinclair (1975). 8 Panitch (1977) reports several cases of the recent collapse of corporatist organizational structure in western Europe.

Political powerlessness and sociodernographic status / 135

1972 “Change in the American electorate.” Pp. 263-337 in A. Campbell and P. Con- verse (eds.), The Human Meaning of So- cial Change. New York: Russell Sage

Easton, David 1975 “A re-assessment of the concept of polit-

ical support.” British Journal of Political Science 5:435-57

Finifter, Ada W. 1970 “Dimensions of political alienation.”

American Political Science Review 64 (June):389-410

Fischer, Claude S. 1975 “Alienation: trying to bridge the

chasm.” British Journal of Sociology 27 (March) : 35-49

Gamson, William 1968 Power and Discontent. Homewood: Dor-

SeY Gilmour, Robert S., and Robert B. Lamb 1975 Political Alienation in Contemporary

America. New York: St Martin’s House, James S., and William M. Mason 1975 “Political alienation in America, 1952-

1968.” American Sociological Review 40 (April): 123-47

Keniston, Kenneth 1965 The Uncommitted. New York: Harcourt,

Kornberg, Allan, Harold D. Clarke, and

1976 “Correlates of regime and partisan ob-

Brace and World

Lawrence LeDuc

ject support in Canada: an empirical examination.” Presented to annual meet- ing of the American Political Science As- sociation in Chicago

LeDuc, Lawrence 1976 “Measuring the sense of political efficacy

in Canada.” Comparative Political Studies 8 (January):490-500

LeDuc, Lawrence, Harold D. Clarke, Jane Jensen, and Jon H. Pammett

1974 “A national sample design.” Canadian Journal of Political Science 7:701-8

Marx, Karl 1975a “Critical notes on the article, ‘The King

of Prussia and Social Reform.”’ Pp. 401-20 in Early Writings. Har- mondsworth and London: Penguin and New Left Review [18441

1975b “On the Jewish question.” Pp. 211-41 in Early Writings. Harmondsworth and London: Penguin and New Left Books [18431

Marx, Karl, and Friedrich Engels 1970 The German Ideology. New York: Inter-

national Publishers [18461 Meszaros, Istvan 1970 Marx’s Theory of Alienation. London:

Merlin Milbrath, Lester W., and M.L. Goel 1977 Political Participation. Chicago: Rand

Miller, Arthur H. 1974 “Political issues and trust in government:

McNally

1964-70.” American Political Science Review 68:951-72

Nie, Norman et al. 1975 Statistical Package for the Social Sci-

ences. Toronto: McGraw-Hill Nielsen, Hans J. 1976 “The uncivic culture: attitudes towards

the political system in Denmark, and vote for the Progress Party 1973-1975.” Scandinavian Political Studies 11: 147-55

Pammett, Jon H., Lawrence LeDuc, Jane Jen- sen, and Harold D. Clarke

1977 “The perception and impact of issues in the 1974 federal election.” Canadian Journal of Political Science 10 (March):93- 126

Panitch, Leo 1977 “The development of corporatism in lib-

eral democracies.” Comparative Political Studies 10 (April):61-90

Phillips, Kevin 1970 The Emerging Republican Majority.

Piven, Francis F., and Richard A. Cloward 1971 Regulating the Poor: The Functions of

Garden City: Anchor

Public Welfare. New York: Random House

Scammon, Richard, and Ben J. Wattenberg 1970 The Real Majority. New York:

Coward-McCann and Geoghegan Schacht, Richard 1970 Alienation. Garden City, N.Y.: Double-

day Schroyer, Trent 1973 The Critique of Domination. New York:

George Braziller Schwartz, Mildred A. 1974 Politics and Territory. Montreal:

McGill-Queen’s University Press Simeon, Richard, and David J. Elkins 1974 “Regional political cultures in Canada.”

Canadian Journal of Political Science 7 (September): 397-437

Sinclair, Peter R. 1975 “Class structure and populist protest: the

case of western Canada.” Canadian Journal of Sociology 1 : 1- 17