This is an open access article distributed under the terms of the Creative Commons
Attribution License (Since 1936, national polls have asked US citizens to self-categorise themselves on a political left-to-right spectrum from ‘liberal’ to ‘conservative’ (Ellis & Stimson, 2012). The measure, however, has historically appeared to be of limited use: party identification is a better indicator of voting intent, and ideological self-placement is only moderately correlated with expert-designated sets of policy preferences (Converse, 1964; Kinder & Kalmoe, 2017). But the study of ideological identity as a driver of American political behaviour has recently re-emerged, this time conceptualised as a social identity that contributes to US political polarisation. Specifically, it is argued that these ideological self-categorisations underlie affective political polarization—the increasing dislike the American left and right have of the other (Iyengar, Sood, & Lelkes, 2012; Mason, 2016). In line with social identity theory (and it’s extension, self-categorisation theory) (Tajfel, 1974; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987), this in-group favouritism and outgroup derogation can be seen as a consequence of citizens’ identification with their ideological identities.
In mid-twentieth century America, the political left and right generally reflected support for, versus opposition to, government intervention, respectively (Downs, 1957). Like other Western democracies, this link between left/right divisions and economic redistribution policy has weakened in the US as post-materialist issues have gained political attention (Dalton, Farrell, & McAllister, 2013; Dalton, Wattenberg, & Wattenberg, 2000; Inglehart, 1997). Over the last fifty years, the left and right in America have increasingly polarised over a broader set of issues (Layman & Carsey, 2002). From the 1960s, liberalism became more saliently associated with the Civil Rights Movement, the Equal Rights Amendment, and Vietnam anti-war protests; social issues continued to become attached to ideological labels with abortion, gay rights, affirmative action, immigration, and gun control creating political divides.
In addition to this extension of political issue associations, the US has seen significant political identity realignment since the 1970s. Not only do liberals today overwhelmingly identify as Democrats, and conservatives as Republicans (Levendusky, 2009), but these identities have also become increasingly homogenous in terms of race, geography, and religion (Mason, 2018b). Social identity theory posits that the convergence and overlapping of social, ideological and party identities (political ‘sorting’) eliminates the ability of identities to ‘cross-cut’ each other and therefore create a more complex and inclusive identity structure (Brewer, 1999; Mason, 2016; Roccas & Brewer, 2002). This political sorting in the US has had the effect of creating an in-group that is increasingly exclusive and important to the individual. The theory also predicts that, when people self-categorise as a member of an important ingroup, they tend to direct positive attitudes and behaviours toward the in-group and negative attitudes and behaviours toward the outgroup. Recent research into the effects of political self-categorisation in the US appear to bear this out. Political self-categorisation predicts political in-group favouritism such as increased activism (Huddy, Mason, & Aarøe, 2015), and support for in-group policy proposals (Malka & Lelkes, 2010). It has also been associated with a number of negative responses to the political outgroup including anger (Mason, 2018a) and distrust (Iyengar & Westwood, 2015), and an unwillingness to engage in discourse with the opposition (Settle & Carlson, 2019; Strickler, 2018). The way in which we think of ourselves politically can even influence our everyday interactions: there is an increasing rarity of close personal relationships in the US across party lines (Iyengar, Konitzer, & Tedin, 2018) and US citizens are more likely to favour their political in-group in a variety of contexts, from online transactions for goods and services (McConnell, Margalit, Malhotra, & Levendusky, 2018), to the awarding of scholarships and jobs (Iyengar & Westwood, 2015; Johnson & Roberto, 2019). Indeed, prejudice against a person of opposing political identity in the US was found to be more prevalent than racial prejudice (Iyengar & Westwood, 2015).
In spite of this evidence regarding the consequences of US political self-categorisation, the understanding of what citizens mean when they self-categorise as a liberal or a conservative remains less clear. These meanings, or identity content according to social identity theory, are integral to understanding the link between self-categorisation and political behaviour (Huddy, 2001): the theory posits that individuals who self-categorise with an ideological group will be driven to behave in accordance with what they believe it means to be a member of that group. The content of identity has been shown to influence attitudes toward the other and possibilities of compromise in intractable conflicts in Israel and Northern Ireland (Bar-Tal, Sharvit, Halperin, & Zafran, 2012; Kelman, 2001; Livingstone & Haslam, 2008). Analogously, the meaning individuals associate with their US ideological identities has the potential to elucidate US political behavior including political polarisation.
US ideological self-categorisations have been analysed on various dimensions over the decades. In particular, significant contributions have been made relating self-categorisations to political issue preferences (Abramowitz, 2010; Converse, 1964; Ellis & Stimson, 2012), social group evaluations (Conover & Feldman, 1981; Ellis & Stimson, 2012) and values (Jacoby, 2014; Peffley & Hurwitz, 1985; Rokeach, 1973; Tetlock, 1986). But little work has addressed these dimensions simultaneously and questions therefore persist as to the structure of ideological identities and the relationship amongst the elements thought to comprise these labels (Feldman, 2013).
The current study seeks to contribute to the scholarship surrounding the content and structure of US political ideological identities by holistically exploring the subjective meaning citizens attribute to liberal and conservative labels. Understanding the way in which those who identify with left and right ideologies construct their in-group identity may provide additional insight into the drivers behind the behaviours attributed to these US political identities in particular, and to political ideological identities more generally.
US Political Ideological Identity as a Social Identity
Social identity theory (Tajfel, 1974; Turner et al., 1987) suggests that the polarising behaviour in today’s US political environment is a natural consequence of political self-categorisation. It posits that self-categorisation creates a social identity that forms part of an individual’s self-concept, leading that person to favour and defend their in-group as an extension of themselves, to adopt the attitudes and beliefs of a prototypical member, and to demonstrate bias against and to derogate the outgroup.
Self-categorisation is the initial step in social identity theory: an individual must see themselves as a member of a social group to be influenced by this membership (Turner et al., 1987). This identification with a social group, such as a political ideological group, causes the individual to act according to what they believe to be that identity’s content or meaning. Social identity content has significant consequences for attitudes and behaviour, the connection has been demonstrated in such diverse studies as those linking racial identity content with academic achievement (Altschul, Oyserman, & Bybee, 2006), US national identity content with attitudes toward immigrants and nuclear armament (Citrin, Haas, Muste, & Reingold, 1994; Kosterman & Feshbach, 1989; Sullivan, Fried, & Dietz, 1992), and the content of Northern Irish Protestant and Catholic identities with aggression toward the other group (Livingstone & Haslam, 2008). How individuals construct the content of their political identities has implications not only for individual behaviour and attitudes, but also for how these groups may be mobilised by party elites, the media, and other ‘entrepreneurs of identity’ (Reicher, Hopkins, Levine, & Rath, 2005). This is particularly true for acquired (as opposed to ascribed) identities as the meaning attributed to the identities will not only impact the consequences of the identity, but also its voluntary adoption (Huddy, 2001).
The content of political ideological identities, like all social identities, are multidimensional and subjective, reflecting an individual’s perception of the norms of the group (Ashmore, Deaux, & McLaughlin-Volpe, 2004). Norms for an ideological identity such as ‘liberal’ or ‘conservative’ may include the personal characteristics, behaviours, and values ascribed to group members, political issues and beliefs, objectives ascribed to the group, and the positioning of the groups in relation to each other and to the nation (Ashmore et al., 2004; Goncalves-Portelinha, Staerkle, & Elcheroth, 2017; van Dijk, 2006). The meaning that individuals attribute to their ideological identities may therefore include, to various degrees, the political issues, social evaluations, and values in the previous research discussed below.
Such a definition of identity content stands in contrast to Converse’s (1964) definition of ideology as having the kind of constraint found in elite discourse. Identity content most certainly reflects the influence of elite discourse, but it may also include a number of other factors. Indeed, Jost, Federico, and Napier (2009) describe ideology as an elective affinity that marries individual differences (personal values, personality traits) and socially created representations of what the labels mean. We therefore employ the broader definition offered by Gerring (1997, p. 980): “Ideology, at the very least, refers to a set of idea-elements that are bound together, that belong to one another in a non-random fashion” as more likely to holistically capture citizens’ perspectives on their ideological identities. To this definition, we add a clarification that political ideology is a belief system that is socially shared (Jost et al., 2009; van Dijk, 2006).
US Political Ideological Identity as Political Issue Preferences
In 1964, Converse published his seminal work concluding that, unlike political elites, the majority of American citizens did not hold a consistent set of liberal or conservative political issue positions—the American public was ‘innocent of ideology’. Fifty years on, Converse’s finding still holds true: self-identification as a liberal or conservative in the US is only moderately correlated with expert-designated left-right issue constraint (Ellis & Stimson, 2012; Kinder & Kalmoe, 2017). This continuing inconsistency between ideological self-identification and citizens’ issue positions led Ellis and Stimson (2012) to draw a distinction between operational ideological identity (the extent to which the issues that citizens support are considered to be liberal or conservative) and symbolic ideological identity (the extent to which individuals see themselves as a liberal or a conservative).
The tracking of operational and symbolic ideological identities over a number of decades reveals a paradox: while a majority of Americans are operationally liberal, only a minority are symbolically so; and although more than 70 percent of self-categorised liberals hold operationally liberal social and economic views, less than 30 percent of self-categorised conservatives hold conservative views on both social and economic dimensions (Ellis & Stimson, 2012). The misalignment between citizens’ self-identified ideology and their support for that ideology’s elite-designated policies underlines the question as to what citizens mean when they identify as a conservative or a liberal if not ‘the’ collection of left or right leaning issues. The content of Americans’ ideological identities—the meaning that they attach to these labels—appears to include elements other than, or in addition to, such policy preferences.
US Political Ideological Identity as Social Evaluation
While operational ideological identity is seen to derive from specific political issues, the meaning attributed to the ‘remaining’ (symbolic) ideological identity has primarily been described in terms of the social groups with which the terms ‘liberal’ and ‘conservative’ are associated (Conover & Feldman, 1981; Ellis & Stimson, 2012; Kerlinger, 1967). Symbolic ideological identification can be thought of as stemming from the social psychological evaluations that individuals hold of groups such as minorities or religious groups that have become linked with either liberal or conservative identities. Such groups represent “dominant cleavages in society” (Conover & Feldman, 1981, p. 643) and liberal and conservative self-identifiers are theorised to evaluate the political identities in light of how they feel about the associated groups.
In an application of this perspective, Ellis and Stimson (2012) theorise that liberalism became increasingly unpopular as the majority of Americans disassociated themselves from the symbols with which this label became linked in the 1960s: blacks, labour unions, and urban unrest. The term ‘liberal’ is much maligned by press and politicians of the right, contributing to this label’s unpopularity. Conversely, the researchers concluded that individuals who describe themselves as conservative may do so due to its accuracy in describing their religious (defined as a strict interpreter of the Bible) or social (measured in terms of a person’s child-rearing views) perspectives. Other work posits that both ideologies are influenced by evaluations of key social groups with labour unions, feminists, and environmentalists more positively evaluated by the left and big business, Christian fundamentalists, and the military by the right (Zschirnt, 2011). Both the positive evaluations of these groups and the hostility toward them were found to be important sources of US liberal and conservative self-identifications.
The Value Basis of US Political Ideological Identity
The structure proposed by Ellis and Stimson (2012) bifurcates ideological identity, acknowledging content contributions from political issue positions and from social evaluations. This dyad does not separately account for the approach that argues that values underpin political ideological identity (Caprara, Schwartz, Capanna, Vecchione, & Barbaranelli, 2006; Rokeach, 1973; Schwartz, Caprara, & Vecchione, 2010). Exploring the relationship between ideological identities and personal values in an Italian sample, Schwartz et al. (2010) employed the widely-accepted Schwartz Values Survey (Schwartz, 1992), finding that liberals and conservatives occupy opposite ends of the personal values circumplex structure, where liberalism is connected with the basic values of universalism, benevolence, and self-direction (collectively the ‘openness’ and ‘self-transcendence’ values), while conservatism is linked to the basic values of security, conformity, and tradition (collectively the ‘conservation’ and ‘self-enhancement’ values).
Political values have also been employed to distinguish between the US left and right. Unlike personal values however, there is no widely accepted set of political values, particularly across countries. In the US, the analysis of the association between ideological identity and political values typically revolves around the putatively core American values such as freedom, equality, and individualism (self-reliance). While both the left and right are associated with the idea of freedom, the left is more typically associated with the value of equality and the right with individualism. In the psychological literature however, these associations have found mixed support amongst researchers according to the method of enquiry. For example, operationalisations of equality demonstrate asymmetry between the two ideologies in their support for the value (Jost, Glaser, Kruglanski, & Sulloway, 2003; Rokeach, 1973). In contrast, other researchers have found a high level of support for the principle of equal treatment on both the American left and right when participants are asked about the value directly (Citrin, Reingold, & Green, 1990; Hanson & O’Dwyer, 2018; Theiss-Morse, 2009). Such inconsistencies suggest a disconnect between how researchers and (particularly conservative) participants define these values—understanding how individuals incorporate these values into their ideological identities may provide insight into this incongruence.
Separate from—but related to—personal values, a distinction has also been made between the types of morals espoused by US liberals and conservatives. Moral foundations theory (Graham, Haidt, & Nosek, 2009) has asserted that US liberals are guided by the morals of fairness and care; conservatives are likewise guided by these two, but are also directed by morals of loyalty, purity, and authority. Fairness and care are considered to be ‘individuating’ morals—morals that focus on individuals as the locus of moral value; while purity, loyalty and authority are considered ‘binding’ morals as they focus on the group (Graham et al., 2011).
The Structure of US Political Ideological Identity
Not only is the content of liberal and conservative identities likely to be multi-dimensional, but the structure may also be asymmetrical. Limited evidence from open-ended questions regarding partisan identity point to possible asymmetry in the content of US liberal and conservative identities. Grossmann and Hopkins (2016) argue that the two US parties are not mirror images of each other; instead, they are different kinds of parties: while the Republican party serves as a vehicle for an ideological movement, the Democratic party is a coalition of social groups. Decades of American National Election Studies (ANES) open-ended question responses regarding what participants like and don’t like about each of the main political parties reveal differing ‘levels of conceptualisation’. Using a method initiated by Converse (1964) and updated most recently with the 2000 ANES (Lewis-Beck, Norpoth, Jacoby, & Weisberg, 2008), the content of participants’ answers to these questions have been categorised according to whether they invoke ideological (relying on a relatively abstract dimension) or ‘group benefits’ (expected treatment of different social groups) talk in their descriptions. In these analyses, there has been a clear difference in how Democrats and Republicans describe the two parties: Republicans have consistently described both parties in terms of ideology while Democrats describe the parties in terms of group benefits (Grossmann & Hopkins, 2015).
Evidence of asymmetrical partisan identity structure was also found when Democrats and Republicans were asked to provide descriptive words that were then sorted into ‘traits’ and ‘groups/issues’, where Democrats used personal trait terms far more than Republicans (Rothschild, Howat, Shafranek, & Busby, 2018). Although political parties may be a more tangible basis for self-categorisation (as a group with designated actors, platforms, and symbols), it is possible that these partisan differences may also be apparent in ideological identities described by the participants in this study.
The Current Study
In the current polarised environment, understanding the content and structure of ideological identities is particularly important. Take for example the bifurcated structure of ideological identity proposed by Ellis and Stimson (2012). This structure has led some to hypothesize that there are two types of political polarisation: affective polarisation and issue-based polarisation (Iyengar et al., 2012; Mason, 2018a). Affective polarisation is attributed to symbolic ideological identification, while issue polarisation is attributed to operational ideological identification. Others, however, maintain that affective polarisation is not a separable phenomenon, being itself centred on issues (Webster & Abramowitz, 2017), thereby integrating the cognitive (issues) and evaluative (affective) elements. Each theory implies a different structure of ideological identifications and source of political disharmony.
The current study aims to more fully apply the social identity framework and contribute to the study of ideological identities by exploring the meaning that those who identify as either liberal or conservative attribute to those US political ideological labels. Through semi-structured qualitative interviews with participants across a wide ideological spectrum, this study identifies how participants constructed their ideological in-groups and outgroups. The goal of the research is to describe the structure and content of these identities from the perspective of the participants and to highlight aspects of identity interpretation and positioning that may have consequences for the study of political behaviour.
Method
This study examined representations of the primary US ideological identities (liberals and conservatives) through semi-structured online synchronous and asynchronous text interviews. A qualitative approach was selected to best capture the complexity and self-reflective nature of these social identities.
Participants
We recruited a sample of 40 participants for this study (20 self-categorised liberals and 20 self-categorised conservatives) using opportunistic and snowballing sampling methods. Although not meant to be a representative sample, we aimed to balance gender and to ensure participation of a variety of generations, geographies, and income levels. Liberal participants were from six different states (California, Iowa, New York, Oregon, Texas, Virginia) and Washington D.C., while conservative participants were from ten different states (Arizona, Florida, Illinois, Iowa, Kansas, New York, North Carolina, Montana, Texas, and Virginia). The two groups were relatively demographically balanced in relation to one another, as indicated in Table 1 below.
Table 1
Participant Demographics, in Number of Participants
| Demographic category | Liberal group | Conservative group |
|---|---|---|
| Gender | ||
| Women | 12 | 7 |
| Men | 8 | 13 |
| Generation | ||
| Silent | 2 | 1 |
| Baby Boomer | 3 | 5 |
| Gen X | 11 | 10 |
| Millennial | 4 | 4 |
| Average age and SD | M = 45.85, SD = 15.49 | M = 48.63, SD = 12.56 |
| Income | ||
| < $50,000 | 2 | 1 |
| $50,000 - $100,000 | 3 | 5 |
| $100,000 - $200,000 | 11 | 10 |
| > $200,000 | 4 | 4 |
| Race | ||
| White | 16 | 19 |
| Hispanic | 3 | 1 |
| Asian | 1 | 0 |
Employing a convenience sample as a basis for understanding perspectives in a population as wide as the US electorate has several limitations. Such a participant group, although geographically and generationally diverse, is not a nationally-representative sample. For example, in comparison to voter turn-out, Gen X voters are over-represented (52% of participants in this study/26% of voters), with the Silent (8% of participants/14% of voters) and Boomer (20% of participants/35% of voters) generations underrepresented by the same (Pew Research Center, 2017); likewise, the participants’ average income is higher than a nationally representative sample. A convenience sample, particularly one in which a high level of motivation is required to complete the study task (the interview), may over- or under-represent certain perspectives.
Potential participants were identified through the first author’s extended contact network, they were solicited initially via e-mail and asked to indicate their interest and informed consent by completing a survey that collected demographic and contact information via a web link. Upon receipt of the survey, the first author contacted the participant to arrange an interview time. No participants were considered to have specialist political knowledge. In partial compensation for their participation, the researcher contributed $10 per participant to one of four charities selected by the participant on the survey. The research received a favourable ethical opinion from the authors’ university.
Procedure and Materials
Seeking to gain a multi-dimensional and person-centred perspective on the content and structure of political ideologies, the study collected qualitative interview data. The aim was to explore participants’ representations of ideological labels by asking them to engage in a certain level of reflexivity about their ideological identity and beliefs. Aware of the current contentious political US environment, we were also keen to employ a format that minimised the threat of direct confrontation and the social pressure of response that exists in a face-to-face interview. Because ideological identities are voluntary social identities created through social interaction, the interview context (as social interaction) has the potential to affect the data collection if the participant would seek to establish social confirmation of their opinions either directly or from the interviewer’s body language. To this end, the interviewer’s ideology was not communicated to the participants directly, and we sought to de-personalise the interview context by conducting interviews via instant messaging platforms (except in four cases where e-mail was used at the participants’ request). The greater visual anonymity in text-based interviewing has been shown to increase self-disclosure and alleviate some of the influence of social desirability on participant response (Joinson, 2001). The contra-argument for anonymity is that such protection may have the effect, as is often seen on social media, of allowing for more extreme views. In addition, text-based interviewing does not allow for subtleties of facial expressions and mannerisms to be collected as data, although these were not considered integral to the aims of this study. The format is also limited in that it requires some level of typing ability and there is little control over participant distractions. To further advance the objectives of comfort and reflexivity, and based on pilot testing of the questions, sample questions were included in the recruiting letter.
The semi-structured interview schedule included approximately 10 open-ended questions that focused on participants’ representations of the in-group and the outgroup. Typical questions were, “what is a liberal?”, “what is a conservative?” and “do you consider yourself to be a typical liberal/conservative?”. By allowing the participants to discuss both their own and the opposing ideology, the interview schedule aimed to capture what participants felt were valued differentiators between the two ideological identities. Interviews were all conducted by the first author and were most commonly completed after 60-75 minutes, although some took up to 2 hours. Most interviews were completed using iMessage and Skype (35), but email (4) and Google Hangout (1) were also used. Two participants completed the interview using their phone, while all others used a laptop to participate.
The data were collected in January and February 2018, one year into Donald Trump’s presidency. Both the Senate and the House were held by the Republicans. Although there were no particular prominent issues in the media during this period, in Trump’s first year he had failed to deliver the Republican healthcare reform bill to repeal the Affordable Healthcare Act (Obamacare), but had removed certain environmental protections, cut taxes for corporations, estates and individuals, and signed an executive order to limit immigration from specific countries for security purposes (the “Muslim ban”). There was little evidence that any particular current issue had an impact on this study.
Data Analysis
The interview transcripts were imported into MAXQDA 2018, a qualitative analysis software application, for organisation and coding. A form of thematic analysis was chosen to explore the data due to its epistemological and analytical flexibility (Braun & Clarke, 2006). An inductive approach was taken, ensuring a bottom-up analysis of the data rather than one driven by particular theoretical objectives. We were, however, alert to the idea-elements that would constitute the content and structure of a social identity including political policy preferences, values, and social evaluations, as well as other items that might comprise a shared representation of the two identities including the positioning taken by participants in relation to their own and the outgroup.
All analyses were conducted by the first author. Prior to initial coding, the data corpus was read and re-read. Initial thematic codes were then generated using a line-by-line approach, ensuring that all of the data were given equal attention. With a view to capturing both the underlying structure of the ideological identity and its content, coding identified both semantic and latent items. In this initial coding, codes were assigned to the entire collection of data, participant by participant. The data were again reviewed by grouping the responses by ideology. This review generated additional codes related to areas of consensus as well as those that distinguished between the ideologies, and the body of data was re-reviewed in light of these additional codes. Codes were then pruned to identify and consolidate themes, and these themes were reviewed based on their relevance to the research question. The themes were then named, defined, described and interpreted. Lexical searches were employed to enhance theme analysis.
Primary themes, including the in-group norms and positioning for each group (liberals and conservatives), are discussed in an integrated fashion below. These themes collectively create a narrative for the content and structure of each ideological identity.
Findings
The purpose of this study was to investigate the content and structure of US liberal and conservative ideological identities through the meaning participants attributed to them. We found that participants constructed their identities through a number of elements: self-attributions, interpretations of the group ideology, and the positioning of their in-group identity in relation to the outgroup and to the nation. The extent to which the descriptions invoked differing types of content—personal attributes, political issues, personal and political values, and representations of the nation—followed certain patterns within the two ideological identifications. The liberal narrative generally revolved around the individual while the conservative narrative most often reflected political ideology and symbols of the nation. Liberal participants constructed their ideological identity from a personal perspective, citing personal values, morals and attributes together with a motivation to progress toward a more equal society. Conversely, conservative participants’ ideological identities were constructed as stemming directly from an American political philosophy, a perspective that equated conservatism and American national identity.
Three main themes were identified in each of the two groups. These themes were Issues make a movement; My politics, myself; and Don’t label me for the liberal group; and It’s Political, I’m with the Group, and Conservatives, to Me, are Really True Americans for the conservative group.
The Liberal Identity: Individuals and Issues
Liberal participants articulated ideological identities that included few specifically political values. Indeed, liberal participants generally resisted categorising themselves as a typical liberal. Descriptions were largely comprised of a collection of political issue positions and personal attributes that centred on concerns for individuals.
Issues Make a Movement
Political values were not primary in liberal participants’ talk about their ideological identities. When asked to define a liberal in this study, only one liberal participant noted the broad principles commonly attributed to US liberalism of either ‘civil rights’ or ‘equal rights’, and the word equality was mentioned by just six of the liberal participants in the whole of the data. That is not to say that these principles were not important to the participants, but that it is not in broad political principles that the liberal participants expressed their ideological identity. They did however often cite a variety of political positions that could be seen as having their purpose in achieving a more equal society. A lexical search indicated that about half of the participants referred to expanding healthcare and the same proportion referenced education in their talk (both longstanding central issues of the US political left); there was otherwise a wide diversity in the issues indicated as core to the liberal ideology, varying from intersectionality (how different types of discrimination interact) to job retraining for the new economy, and from Native American issues to the environment.
While the liberal participants appeared to see themselves as proponents of specific issues; their link between these issues and a national political ideology was rarely articulated, and perhaps even resisted. Participant 7L described her position as follows:
The liberal ideology itself is not important to me, because it is the SYSTEM of ideas and ideals. I look at each individual issue rather than the whole system.
This talk clearly puts issues ahead of a stated ideology. In fact, “what defines a liberal will inherently change because of the progressive nature of the beliefs” (Participant 5L). This response abrogates an opportunity to define ‘liberal’ in terms of core values, but instead links the definition to progress generally, presumably linked to support for certain issues. This statement also suggests that the heterogeneous issues may be connected in a common vision of the strategic project that is broader than national political ideology—progress toward a better world, without impediment to individual expression: “People just need to be able to be who they are. Without judgements.” (Participant 4L), “I think itʼs wanting to make sure all people have what they need to be the best version of themselves.” (Participant 10L). This talk speaks to a value of freedom—freedom for citizens to realize their true self through a greater degree of equality. Rokeach (1973) proposed that both liberals and conservatives hold freedom in high regard, while liberals also highly value equality. Building on this, Tetlock (1983, 1986), proposed that holding these two conflicting values drives the more complex decision-making and communications of the left. The talk in this study demonstrates a process that liberals may use to reconcile these values: by defining freedom not as ‘freedom from government intervention’ (as is typical of conservatives) but as ‘freedom of expression’, and providing equality as the means to provide this freedom.
These representations of shared vision provide little expressed evidence of a commonly employed national political doctrine, common phrases or vernacular in liberal participants’ discussion of their ideological identity. Liberal participants positioned themselves as moving toward a shared representation of a better way of life, while specific political issues were conceptualised as the milestones that are to be achieved in moving toward this ideal. It is possible that this heretofore unknown world state is difficult to visualise and agree upon, and therefore remains unarticulated. Alternatively, a persistent absence of political ideological talk in American left-leaning discourse may be intentional—undertaken both to avoid ideological conflict with self-categorised conservatives by focusing on the liberal issues that appeal to this group as operational liberals and to appeal to liberals’ group interests (Grossmann & Hopkins, 2016).
My Politics, Myself
In addition to issue positions, liberal identifiers often described their ideological label in terms of personal values and behaviours. The extent to which these elements were equated with being a liberal was notable not only because of its semantic consistency within the liberal participant group, but for its latent meaning. This talk was often in response to the prompt, “What is a liberal?” (a question ostensibly about political ideology, not personal attributes). By invoking personal attributes in response to this question, the participants have brought forth the importance of these elements and have positioned the individual at the centre of the identity. This structure indicates a ‘bottom-up’ influence on political beliefs: the individual, not a prescriptive group ideology, was seen to inform liberal political values.
Participants positioned the US liberal in-group as open, caring and outward-looking; they saw themselves as self- and societal- improvers, “seeking to better themselves and society” (Participant 10L). Consistent with research that correlates measures of the value of openness with liberal self-identification (Schwartz et al., 2010), being ‘open’ was seen by participants as core to the American liberal identity. Participant 17L connected his personal outlook to being a liberal in this way:
I know I enjoy a variety of cultures, a variety of people, I still love to learn and explore, and I think that is the basis of who I am. I find humor in the absurd and as my mother always said it is better to laugh than cry. I seek to find solutions and don't see most things in life as insurmountable but instead to find ways to solve problems and meet people part way without either having to capitulate.
This multifaceted description has openness at its core, defined both as tolerance and as learning. Such a definition reflects the personal values of self-direction and universalism between which liberal political values were found to be positioned in an Italian sample (Schwartz et al., 2010). Openness was most often cited as intellectual curiosity, as described by Participant 8L:
For me, a liberal equates to being open to an array of ideas and perspectives. Continually exposing yourself to new ideas, ways of thinking, etc., with the express intent of broadening your perspective. Challenging yourself to avoid the trap of egocentricity by tapping into views that may differ from your own, experiencing different cultures and being open to adapting, continually seeking new information/learning, etc.
Participants frequently discussed the critical evaluation that openness affords. This positioning has the effect of validating not only the attribute of openness (because it allows for better reasoning), but it also lends validity to the liberal issues that are the outcome of critical thinking. Importantly, it sets critical-thinking liberals against conservatives who blindly follow ideology. Frequent references were made to the perception that American liberals are “educated” – not necessarily formally, but in the sense of being informed on historical or political issues. Liberal participants positioned themselves as thinkers who are hungry for information, and as being in control of their ideas.
I think a liberal is someone who is balanced, uses their intellect, values that consider the best outcomes for making life decisions and is open to possibilities that go beyond ones (sic) own limiting life experience. I think openness is the key. (Participant 17L)
Here, we again note that political ideas and behaviours are driven by the individual.
Conservatives, on the other hand, were seen by liberal participants as either selfish or as victims of their upbringing, their religion, and their geography – factors that were seen to keep conservatives from the advantages of experience or education that would open them to more liberal ideas. Conservatives were seen as “crazy uninformed” (Participant 12L), and as having been duped by those in power and their doctrines.
And, honestly, I really donʼt think itʼs 100% their fault; when you trust the wrong politicians or the wrong religious doctrines or believe incorrect facts and figures, this is what happens. Theyʼre scared and angry because people have promised them things or told them things arenʼt true. They donʼt understand that the people they trust are actually the people that are making money off of the lies they tell them. They donʼt understand that they are being kept where they are because it benefits the people who put them there and are keeping them there. Itʼs actually pretty sad, really. (Participant 10L)
This positioning by liberal participants of US conservatives as being blinded by socially constructed belief systems is set in contrast to the liberal individual-driven beliefs. The talk cast conservatives as either allowing themselves to be dictated to by societal constructs or as victims of their environment and of elites, setting this against liberals’ self-positioning as the more aware, outward-looking, and personally in control of their values. Attributing conservatives’ positions to their environment also serves to invalidate conservative positions while allowing liberals to not violate their value of openness to all perspectives. This observation also offers an interesting comment on the ‘ideo-attribution effect’ phenomenon wherein liberals have been noted to make situational attributions for social problems, while conservatives tend to make personal attributions (e.g. Sniderman, Hagen, Tetlock, & Brady, 1986). Liberals’ tendency to ascribe conditions such as poverty and unemployment to environmental causes has been attributed to reasoning motivated by the need to eliminate the cognitive dissonance that personal responsibility for one’s social or economic woes would create for liberal egalitarian values (Skitka, Mullen, Griffin, Hutchinson, & Chamberlin, 2002). For the liberal participants in this study, there should be no such value conflict related to the idea that conservatives may be responsible for their own political attitudes—yet the situational attribution persisted. This continuity of attribution type where there is little political ideological value conflict suggests a driver not confined to the political realm.
Don’t Label Me
The importance of individuality to the US liberal identity also came through in liberal participants’ assessments of their prototypicality. Of the 20 participants who identified as liberal in this study, only four gave an unqualified “yes” when asked if they were a typical liberal. Generally, participants expressed an uneasiness with defining typical members of the group (e.g. “Is there a typical liberal? [Participant 7L]). Participants perceived the group as having such diversity that a prototype was impossible to imagine: one-quarter of the liberal participants indicated that they weren’t sure what a typical liberal was or if one actually existed. This perceived ingroup heterogeneity (as compared to the conservative participant group, described below) again speaks to a greater personal, over group, identity (Brewer, 1993). Participants commonly qualified their typicality for their geography, age, or social group:
HA! I don't consider myself too typical in any way. But among the people that I live around in NYC and identify with I would be somewhat typical. As part of a larger nation, less typical. (Participant 3L)
Other participants qualified their typicality for their level of political sophistication: being “more pro-business” (Participant 1L), “more informed” (Participant 10L), “further along on the advancement of social issues” (Participant 3L), “more aware” (Participant 18L) or more “active” and “intersectional” (Participant 12L) than most. Again, none of these qualifications revolve around a discrepancy in beliefs, they are personal attributes, behaviours, and environmental influences, consistent with the higher level of trait responses found in Democrats (Rothschild et al., 2018). This positioning both speaks to the perception of the American liberal group as a collection of individuals rather than a group of shared national political ideology; it also supports Ellis and Stimson’s (2012) assertion that there is a reluctance to identify with the liberal label in the US—even those who do embrace this label tend to deny typicality. Ellis and Stimson attribute the reluctance to identify as liberal to the negative symbols of blacks, unions and urban unrest associated with the label, but our work offers an additional explanation regarding the incongruences between operational and symbolic ideological identities. The persistent theme of individual expression and the condemnation of blind acceptance in liberal participants’ talk raises the possibility that their political identity is seen as a personally derived set of issue positions. Such an individual identity construction might defy ascription to a pre-ordained set of political values or beliefs and could be more resistant to political ideological labelling of any kind regardless of the symbolic associations.
The relationship between liberal participants and their political labels was further illuminated in talk regarding their related political party. Although some participants saw the Democratic party as too slow and subject to the corrupting influences of power and money, most participants cited little difference between being a liberal and being a Democrat, noting that, although not highly aligned with the participants’ beliefs, the party is the “only game in town” (Participant 11L).
There really isn't much of a choice at this juncture to be other than democrat if you are a thinking human being with a concern for the welfare of others. My moderate tendencies are to take people as they are and help them move forward without hard line absolutes. For me the democrat thing is by process of elimination and the independents are often coming from an unrealistic place that has no hopes of accomplishing something. (Participant 17L)
Together, these observations position both the liberal and Democratic identities as simply “best fit” conduits for what these participants generally described as a personal ideological identity. In sum, liberal participants appeared to conceptualise their identity first and foremost as a confederation of individuals who possess particular personality attributes, personal values and a vision of a better, more equal, world where individuals are able to fully express their personalities and talents. Political ideology was positioned as the result, not the driver, of these representations.
The Conservative Identity: Ideology and The Nation
Unlike liberal participants in this study, conservative participants saw themselves as typical group members and clearly articulated their ideological group’s political beliefs and goals—a system closely linked to the nation.
It’s Political
The conservatives in this study characterised their ideological identity as consisting of a defined set of national political values: limited government, adherence to the constitution, and self-reliance. As Participant 17C concisely expressed, a conservative is…
An individual who respects the constitution of the United States, believes in a society where if you work hard you are allowed to keep the gains of your efforts and you are not unnecessarily burdened by federal or state interference.
In contrast to liberal participants’ ‘personal values’ definition of their group, these beliefs are clearly political: Participant 17C refers to the US, to society, and to federal and state governments. The participants’ most common directly noted political belief was limited government, characterised as a founding and Constitutional American value (the alignment of the conservative and American identities is discussed in the section below).
Although not generally referred to directly when prompted to define a conservative, woven throughout participants’ talk was the central and pervasive belief uniting the conservative identity: self-reliance.
I believe in personal responsibility. First and foremost get your own house in order before you worry about anybody else’s house. (Participant 7C)
Self-reliance was seen to be demonstrated at an individual level with personal fiscal responsibility and a sound work ethic. Participant 20C explains the personal importance:
As I grew older and had kids I more and more thought about what things would be like for them when I'm gone from this earth. It became more evident to me that the work ethic that my parents and grandparents instilled in me was THE most important gift they ever gave me. It is the key to success. I wanted my children to know and understand that they were the masters of their own destiny and did not need the government to help succeed. If they worked hard enough they could do/become whatever they desired.
Similar to the liberal participants’ valued traits of openness and tolerance, self-reliance was a behaviour that conservative participants value in themselves and expect of others and of their country. Participant 1C illustrates this by contrasting herself and a colleague:
I worked with a guy named Matt. I was talking with he and his wife at a dinner one night. They just found out they were expecting and had already decided his wife was not going back to work after having the baby. I was surprised. I knew that they could not live on one salary. Neither of them went to college. He said he couldn't afford it and he didn't want to have tons of loans to pay back. I explained that I went to college and my parents didn't pay for anything. I worked all through school and had financial loans. It took me about five years to pay off the loans but I had a degree that allowed me to have a career instead of just a job. Yes it took time and it was very hard at times but it was worth it. They both said that was stupid and that they could get jobs anytime they wanted. I in a very nice way asked how they would manage with one income. They said that they could get assistance because he only made so much and that she wasn't working. They were only 23 yrs old. I couldn’t understand why they were married and having a child when they could not afford to live on their own. This is not what they were thinking?????
This passage supports the narrative of benefits recipients choosing to rely on the state for support, a scenario that is in direct opposition to the US conservative principle of self-reliance. Participant 1C hints at an argument put forth by a number of other conservative participants: assistance programs degrade individual character and society by undermining self-reliance. The programmes were variously characterised as supporting “a destructive lifestyle" (Participant 1C), noting that “recipients never learn to fend for themselves” (Participant 13C), as “detracting from that [work] ethic and reducing the individuals (sic) feeling of self-worth” (Participant 20C), and as taking away from our children the “opportunity to succeed or fail or to make it on their own” (Participant 9C). This line of argument positions US liberal policy as robbing current citizens and future generations of an important personal characteristic that is seen as the key to both personal and national success. This position takes the welfare debate beyond a simple ‘free-rider’ issue to a discussion about personal and national character.
I’m With the Group
The majority of conservative participants considered themselves to be typical without qualification. In participants’ descriptions, there was a clear prototype of a conservative. If participants did not consider themselves to be prototypical, the reasons they cited were differences in political issues. Notably, the majority of conservative participants, without prompting, indicated that they considered themselves to be socially moderate/liberal, most often citing being “ok with gay marriage” (Participant 4C), but there were also single instances of varying from prototype on abortion (Participant 6C) and by being irreligious (Participant 17C). None of these positions were perceived to be in conflict with seeing themselves as a typical conservative. Such a definition stands in contrast to what is generally regarded as conservative values, including those positions used to assess operational ideological identity. Not unusually—yet in contrast to the liberal participant construction—the conservative ideology was constructed as a social group defined primarily by members’ beliefs in the conservative political philosophy.
Ellis and Stimson (2012) describe self-identified conservatives who hold liberal political positions ‘conflicted conservatives'. The researchers attribute this conservative peculiarity to self-interest: in spite of their ascribed ideological identity of limited government spending, citizens prefer liberal policies that confer benefits and services (e.g. Medicare) when they are asked to make decisions on a policy-by-policy basis. Our analysis indicates that the phenomenon may not be limited to a policy-by-policy context. Participants' conscious acknowledgement of their liberal social stances did not keep them from identifying as a typical conservative, and their fiscal concerns revolved around fiscal responsibility, not simply less spending. This inconsistency not only attests to the strength of the conservative group identification, but also raises a question regarding the operational definition of generally accepted ‘conservative' positions. If the majority of conservatives don't hold conservative policy positions, perhaps the positions considered to be conservative by political scientists (often based on the willingness of citizens to spend more or less government money) do not reflect those considered to be conservative by the general public. After all, political ideology is a social representation, an object constructed from public opinion (Jost et al., 2009); what defines conservative positions should therefore reflect shared opinion—not simply the views of political experts—if we are to advance an understanding of US conservatism. This area requires additional work to more fully understand this incongruence, but our work suggests that there may be space in the content of the conservative identity for what is currently considered to be liberal policy.
Unlike their ideological group membership, there were a significant number of conservative participants who distanced themselves from the Republican party. In line with conservatives’ traditional support of limited government, this reluctance to identify with the party appeared to be primarily due to participants’ distaste for politicians generally, noting that they felt closer to their ideology than to their party. Participant 1C noted “I used to consider myself a republican but feel that the word is more about power than it is about what is best for our nation”. Such claims are consistent with the right’s traditional distrust of government, but they may also be due in part to participants wishing to distance themselves from unpopular actions of the current Republican president. However, Participant 15C put it this way: “Being a conservative is about principles. Being a Republican is about policy”, where principles are seen as closer to the identity of the right.
Conversely, there was also evidence that the Republican identity may be more salient for some. Participant 10C, when asked to define a conservative, was confronted with the idea that she did not meet her own definition as “trying to uphold the traditional values and religion” because she “support[s] gay marriage, etc.”. She worked out the inconsistency, unprompted, as follows:
Participant 10C: “republicans are conservatives”
Interviewer: “Yes, that’s generally true. Are you saying that you are a Republican?”
Participant 10C: “I’m a republican..”
Interviewer: “And how do you think conservatives and Republicans differ?”
Participant 10C: “I don’t think they do. that’s the reason i say I’m a conservative bc I’m a Republican”
This priority of party over ideology was mirrored in the priority of candidate over ideology that the participant later noted in her family’s political behaviour:
so I'm from a very small town in South Texas. my sister's and my brother vote on whoever my mother tells then to vote for. my mother votes for whoever this person knows.
Together, this contrasting identity priority illustrates the significant influence of social influence on political identification, not only on whether citizens identify on the left or right (Campbell, Converse, Miller, & Stokes, 1960), but also on the priority of these political identities.
“Conservatives, to me, are really true Americans” (Participant 3C)
Conservative beliefs, as communicated by the study’s participants, were commonly equated with American values, success, and strength; a number of conservative participants positioned themselves as the defenders of American political philosophy, while liberals were often equated with or seen as “heading toward” (Participant 18C) socialism: “Today's liberal in the USA reminds me of what I would call socialism.” (Participant 9C). By assigning liberals this political ideology, conservative participants both carry over the strong ideological framing of their in-group to discuss the outgroup and position the outgroup as un-American.
The left has gone so far now that its (sic) as if they don't understand our Constitution or the principles of Capitalism. They, for some bizarre, ideological reason, believe Venezuela or Cuba are a better model. They crave socialism. They live in the wrong country if that is what they want. The "left' (liberals) in our country, at this point, are pushing socialism/ communism with thought control, speech control, etc. (Participant 8C)
Socialism as a threat was mentioned specifically by seven of the conservative participants (note that no liberal participants mentioned the word socialism), indicating that it was an entire ideology, not just particular issues or values, that was perceived as a threat. By constructing the US political conflict as a battle between two ideologies (American and non-American), conservatives firmly align themselves as the defenders of the country. Participant 3C described the conflict as follows:
our society has been infected with socialism and entitlement. those that are holding true to American values are fighting back. we need to remind ourselves what it means to be an American. and those that can’t support who we are need to either move or frankly shut up/back off. this is our constitution and who we are. those socialist views don't belong here. we need to eradicate socialism.
This talk directly links the US conservative self to the nation, with conservative political beliefs at the epicentre. Conservative values were conceptualised, not as an independent political philosophy, but as the same as the American founding philosophy.
Positioning conservative and American political philosophy as one in the same allows conservatives to gain a moral high ground as defenders of the nation against the invading philosophy, it increases the importance of fighting for conservative values against the un-American liberal aggressor. The association of socialism with liberalism is clearly embedded in the conservative discussion. The anti-American connotations of this ideology may offer another reason for operational liberals to distance themselves from the liberal label: Democrats as well as Republicans highly value their American identities (Huddy & Khatib, 2007).
Conservative participants’ American identity was an important social identity for the conservative participants and talk linked conservative principles to the country’s perceived strength, position in the world, and success:
It is more about creating a country that you are proud of and you want to be a part of. I think conservative values allow the United States to continue down a similar path of those who founded our country. In my opinion it will make us stronger as a country than if we were to adopt more liberal principles. like when the United States is seen as a world power who will help those who need it, but will still stand up against those whose values are in direct opposition to ours. (Participant 6C)
Conservative values were also credited as providing the political philosophy for allowing for “a strong foundation for innovation” (Participant 20C). America’s past success was often attributed to the “freedom” afforded by conservative policy:
a while ago, i did some research on inventors. I found that most of the great inventors came from America. I find that fascinating. it means, to me, that in a truly free society, you have the ability to create and rise above all else. and, society will benefit great inventors don't come from places like North Korea....which means someone is giving them the technology (Participant 3C)
It is in talk of this type, citing past success as a validation of conservative values, that conservative participants’ reverence for what they see as their conservative/American political philosophy becomes apparent. American history was seen as “something that should be honored and revered not changed for the sake of change or change because itʼs of popular opinion or simple fallen out of favor” (Participant 7C). This attribution of liberal change as short-sighted and as lacking in respect for the nation was common and is consistent with positioning conservatives as the defenders of the nation. Most conservative participants discussed US liberals as naïve:
[A liberal is] an individual who really doesn't understand the cause and affect (sic) of individuals actions. Is not realistic with their ideals and the entire population. (Participant 1C)
They were generally seen as having little awareness of the threat their policies pose to the nation and to the freedoms of other Americans: they were seen as well-intentioned but ignorant.
Conservative participants appeared to revere and to identify with the concept of the American nation, its political philosophy and its strength and position in the world. By self-identifying as conservatives, they are also identifying as defenders of the American political philosophy. This pervasive alignment speaks to a value that moves beyond, and may operate at a different level than, the traditional political values of freedom, equality, and individualism. Reverence for the nation was a key component of the conservative ideological identity. This finding is in line with the moral foundations theory of conservatives tending to support group-enhancing morals (Graham et al., 2009). The group, in this case, being the nation. Indeed, in a geometric model of individual value choices, Jacoby (2014) found that support for patriotism was a key differentiator between liberals and conservatives. Further, this group-enhancing value stood in opposition to the individual-oriented values of equality and freedom.
Conservative participants’ ideological adherence enabled them to readily envision a prototypical member and to provide clear and consistent meaning for the conservative identity. This engagement with the US conservative identity suggests that this group would lend itself more easily to mobilisation. In addition, with national identity being one of the most accessible and powerful identities for mobilisation (Billig, 1995), the alignment with national values gives conservatives and the Republican party access to a social identity that is highly and chronically salient. By ‘owning’ American values, conservatives are able to easily cast liberals as the un-American outgroup, and associating conservativism with the country’s strength implies that alternative policies may be a threat to this strength. This narrative offers the conservative identity as a means by which citizens may express their support for American values. Not only does this positioning give an advantage during settled times, but any external threat to the country is likely to result in citizens moving toward conservativism as a way to express their American identity, such as in the period following September 11, 2001 (Li & Brewer, 2004).
The conservative participants’ simpler messaging is consistent with previous research (e.g. Tetlock, 1983) and has been associated with a less complex style of issue evaluation by conservatives (Tetlock, 1986). From a group perspective, this nationally-linked, consistent messaging also promotes unification of the US conservative group and stands in contrast to liberal messaging. The conservative participants’ association between conservative and American values faced little counter-narrative from liberal participants, a pattern that exists in the wider public (Gidron, 2018). The lack of ideological talk provides no counter for conservative discussion: US liberals may have little national or political ideological language with which to engage conservatives in what conservatives might feel is the central debate. While conservative participants appeared to prefer to discuss political values, liberals were more likely to focus on issues. When no ideological counter-narrative is offered by liberals, those on the political right may fill this void—as the conservative participants in this study did—with their own choice of ideological vernacular: socialism.
Conclusion
Liberal and conservative participants’ construction of their ideological identities differed in both structure and content. Liberal participants’ talk centred on the individual, both in terms of the individual values and characteristics that define the liberal identity and in the concerns for individual freedoms (defined as personal expression). They represented themselves as a highly diversified group that eluded prototypicality, often eschewing the ‘typical liberal’ label, but they found commonality in personal values and the shared ambition of a better world. Conservative participants were more apt to embrace their ideological label and to discuss group norms and concerns in ideological terms, both political and national. They positioned themselves as defenders of the nation against the threat of socialism and the weakness of character that it engenders. These participants closely linked their personal, political and national identities through the thread of self-reliance – seen as the key to the nation’s past and future success – and reverence for the nation. In both groups, individual freedom was paramount: liberal participants extolled the need for individual expression, while conservative participants valued freedom from government interference.
The elements employed by liberal and conservative participants to construct their ideological identities were asymmetrical. Where liberal participants used political issues and a future world vision, conservative participants employed national political ideology. Supporting and extending the two ‘levels of conceptualisation’ originally identified by Converse (1964) (the tendency for Republicans to describe parties in ideological terms and for Democrats to use group interest references), and recently discussed by Grossmann and Hopkins (2016), our findings suggest that US ideological identities are multi-faceted structures in which liberals and conservatives employ political issue preferences, national political ideology, and world vision asymmetrically. Further, these differing levels of conception mirrored representations of different group concerns. Liberal participants’ discussion of individual rights and oppressed groups contrasted with conservative ideological talk that aligned itself with American values—concerns related to individuals versus concerns related to the nation—a division similar to the contrast between individuating and binding moral values (Graham et al., 2009). These identified asymmetries highlight the identity elements that may be most salient for group mobilisation, conflict, and political communication, suggesting a number of avenues for future research.
The ideological/individual centres of concern framed the threat each group felt from the other. According to liberal participants, the threat posed by conservatives is to individual expression and is caused by conservatives’ blind devotion to their political, national, or religious ideology. The threat posed by liberals, according to conservative participants, is centred on their support of the un-American ideology of socialism, cast as a threat to America’s strength and character, undermining the key value of self-reliance and contrary to the founding values of the nation. These representations of the outgroup may be used to stir intergroup hostility in political communications, fuelling polarisation and jeopardising policy negotiations. Being alert to the source of an outgroup’s concerns and their perceptions of one’s in-group (regardless of the perception’s accuracy) can create the space that allows a group to create an alternative narrative. For example, the current study highlighted the narrative of ‘assistance programmes undermine the American value of self-reliance’ in conservative participants’ talk. Framing assistance programmes as a nation-enhancing may offer more opportunities for cross-party communication by offering a platform from which to discuss what is an important identity for the conservative group. Future experimental work could explore the impact of messaging that differentially appeals to these facets.
Although the study reflects perspectives on ideological identity from a geographically and generationally-diverse group of participants, it is not a fully representative of the electorate. We may therefore be missing certain perspectives on the content of these ideologies, and the findings cannot be generalised. Our work may however serve as the foundation for future work surrounding the conceptualisation and consequences of American and other left/right ideological identities. With the increasing prominence of non-economic issues and the decline of political party attachments, the meaning and role of left and right political ideological identity is changing, not just in the US, but across affluent democracies. Although the current study is specific to the US context, the general finding of asymmetrical identity content structure (of individual or group ideological perspectives) between the political left and right may provide useful means by which to discuss other political ideological and party identities throughout the world. In addition, exploration of the significant role of national reverence on the US right and the relationship between this group-enhancing position and the traditional equality and freedom values of liberal democracy may illuminate identity structures related to populist movements in other geographies.
Our study provides evidence that US ideological identities are multi-dimensional structures in which liberals and conservatives to differing degrees embrace concerns for individuals and for ideologies, differentially employing social representations of political policy issues, personal and political values, national ideology and world vision. This work adds to the growing body of research that employs social identity theory to conceptualise political polarisation; it elaborates on the content of ideological identities while also contributing to the study of ideological asymmetries. Conceptualising ideological identities as different types of identity may not only inform social psychological models of affective political polarisation by providing insight into the behaviours that are the consequence of ideological identification, but may also offer new strategies for political communication and policy negotiation. Awareness of these contrasting identity constructions and understanding that the two groups may be advocating different, but not inherently opposite, elements allows more space for compromise by identifying areas that may be most available for negotiation, a step toward reducing US political polarisation.