header image

Volume 25, No. 1, Art. 7 – January 2024

Doorways of Understanding: A Generative Metaphor Analysis

Katie Haus, Jaclyn Hadfield, Kathryn J. LaRoche, Barbara Dennis,
Ronna C. Turner, Brandon L. Crawford, Wen-Juo Lo & Kristen N. Jozkowski

Abstract: In this paper, we explore the use of a generative metaphor for analyzing qualitative interviews on abortion attitudes. U.S. abortion attitudes are notably complex and multidimensional, thus, requiring subtle, complex, and multidimensional tools of study. We used the generative metaphor of a "doorway" as an analytic tool to enable new understandings of abortion attitudes as expressed across 24 one-on-one semi-structured qualitative interviews with U.S. adults. The doorway metaphor gave us an understanding of the ways in which participants thought of their abortion attitudes as open to revision or change to some degree while also being closed to revision in other ways. This spectrum of openness and closedness does not come into view when examining abortion attitudes through the dichotomous framings. In this methodological paper, we thoroughly describe how we used the metaphor to explicate the complexities and multi-dimensionalities of a person's abortion attitudes.

Key words: generative metaphor; abortion attitudes; analytic metaphor

Table of Contents

1. Introduction

2. Generative Metaphors as Methodological Tools

2.1 The empirical use of metaphors to study abortion attitudes in qualitative research

2.2 From conceptual metaphors to generative metaphors

2.3 Using generative metaphors as a qualitative data analysis tool for understanding abortion attitudes

3. Methods

3.1 Participant sample

3.2 Methods of data collection and analysis

3.3 Validity

4. How the Metaphor Worked Analytically

4.1 The components—A detailed description

4.2 Doorway: An example of the generative metaphor as an analytic tool

5. Discussion and Conclusion


Appendix: Interviewee Pseudonym and Demographic Characteristics






1. Introduction

Through qualitative studies of abortion attitudes researchers have been able to better articulate the complexities we find oversimplified in political discourse (BRUCE, 2020). Understanding complexities in abortion attitudes could foster opportunities for public debate to move beyond polarized labels and identifications. In this article, we focus on the development of a generative metaphor—a metaphor that enables new ways of understanding phenomena (as introduced by SCHÖN in 1979). We used the metaphor "doorway" as an analytic tool for better understanding complex abortion attitudes amongst 24 interviewees participating in the qualitative portion of a larger multi-method study of abortion attitudes. SCHÖN developed the concept "generative metaphor" as a way of moving policy development conversations toward new "perceptions, explanations and inventions" (p.142). We further this concept of generative metaphor by demonstrating its use for qualitative data analysis and illustrating how it creates new understandings. [1]

Often, researchers leave implicit the metaphors they use for their own analysis. Our aim in this paper, however, is to demonstrate the methodological practices associated with using a generative metaphor to study complex attitudes in a reflexive, inclusive and controlled way. We found that the doorway metaphor enabled us to articulate ways of thinking about abortion attitudes that were more subtle than the commonly assumed framings pro-choice/pro-life, moral/legal and religious/feminist (LEE, McLEOD & SHAH, 2008) which are known to erase abortion attitude complexities. SCHÖN (1979) argued that new understandings produced through the use of a generative metaphor are "likely to bring us into sharper and more explicit confrontation with frame conflicts" (p.150) such as those fabricated through contemporary discourse binaries regarding abortion attitudes. For example, by exercising the doorway metaphor, we gained an understanding of the ways in which participants thought of their abortion attitudes as open to revision or change to some degree while also being closed to revision in other ways. This spectrum of openness and closedness does not come into view when examining abortion attitudes through dichotomous framings. In this paper, we intend to build an understanding of a generative metaphor as a methodological tool. We describe how the generative metaphor 1. emerged in our analysis, 2. was taken up as an analytic tool, and 3. offered an opportunity for us to restructure and reframe our understandings/analyses of complex moral/legal phenomena in ways that are applicable to other qualitative research on similarly complex attitudinal phenomena. [2]

In the study of attitudes, there has been a tenuous connection between the attitudes one espouses, for example on a survey, and how one actually behaves (GARRETT, 2010). Such weak relationships must be methodologically confronted when researchers are interested in understanding the complexities of human attitudes. We argue that the "doorway" metaphor made it possible to reconstruct the ways in which participants situated themselves in relation to their abortion attitudes rather than focusing solely on the attitude being claimed. Specifically, our analysis explicated the aspects of interviewees' abortion attitudes that they attributed to stability (i.e., aspects they held firm to). Additionally, and contrastingly, we are able to report on the detailed aspects of participants' abortion attitudes that they indicated were open to alternative perspectives and ways of acting. [3]

We organized this paper into three main sections. We begin with a review of the empirical and theoretical literature relevant to our project of using generative metaphors as methodological tools (Section 2). We follow this with a description of our methods through which we provide details of the current study with a focus on the steps we used to analyze the data with a generative metaphor (Section 3). In the main body of the paper, we report on how the metaphor worked analytically (Section 4). First, we describe the components of the generative metaphor (Section 4.1), then we demonstrate how the metaphor was applied by illustrating its use in one interview (Section 4.2). In our conclusion we argue that the generative metaphor "doorway" is a useful analytic tool for the study of complex social issues and the attitudes people hold in relation to those issues (Section 5). [4]

2. Generative Metaphors as Methodological Tools

In this section of the paper, we 1. review the use of metaphors in the study of attitudes, 2. illustrate the theoretical path linking two specific ways of thinking about metaphors to the use of metaphor as a methodological tool, and 3. situate the tool in the study of abortion attitudes. Our review of the empirical literature indicates that qualitative researchers have primarily engaged in an analysis of metaphors that were explicitly used by participants (CAMERON, 2003). In contrast, we used a generative metaphor in a reflexive and controlled way to explicate what participants indicated about their abortion attitudes. This matters because we know that research on attitudes cannot stop at the explicit claims people make about their own attitudes (GARRETT, 2010). Our argument is that researchers can use generative metaphors as qualitative analytic tools to better study complex attitudes. One of the reasons this was so useful for our study is because the "doorway" reflected how the interviewees situated themselves in relation to what they were saying about their attitudes. We specify how our work has moved from "conceptual metaphors" toward "generative metaphors." The main claim of conceptual metaphor theories is that we think of one domain (conceptualizing our own abortion attitudes) in terms of another domain (the "doorway" metaphor). Generative metaphor adds the possibility of thinking anew by seeing the phenomena of interest from a different vantage point. [5]

2.1 The empirical use of metaphors to study abortion attitudes in qualitative research

There is a prolific history of scholars analyzing metaphors in the study of literary texts, but their application by researchers in field-based studies is less common. Mostly, metaphors have been used in the study of attitudes by 1. inferring attitudes through participants' uses of metaphor (CAMERON, 2003), 2. reconstructing an underlying organizational structure of attitudes through metaphors (FURUNES & MYKLETUN, 2007), and 3. presenting findings through metaphors (SATTERFIELD & GODFREY, 2011; SMITHIES & LATHER, 1997; STEELE, BAIRD & DAVIES, 2022). [6]

The most typical way researchers have conducted an analysis of metaphors in attitude studies has involved articulating metaphors that participants directly expressed themselves (CAMERON, 2003; WHEATLEY & VATONEY, 2022). CAMERON (2003) found that researchers can learn important information about speakers' attitudes, values, and ideas through the metaphors they invoke. HERACLEOUS and JACOBS (2008) argued that people's political thinking is aided through metaphors which can be reconstructed from how they express their political perspectives. [7]

BRUGMAN, BURGERS and VIS (2019) presented evidence that conceptual metaphors are linked with one's thinking. They explored the link between metaphors and thinking by comparing political conceptualizations of "metaphorical understandings of an issue under investigation" (p.46) with metaphorical expressions used by participants to invoke their politics by talking about something else (ibid.). BRUGMAN et al. considered these as two different levels of analysis of metaphors—the word level and the conceptual level. They found that conceptual metaphors had a more significant impact on how people were thinking through political issues. Though they did not look specifically at abortion attitudes, their conclusions about metaphors are relevant to our study because we are interested in understanding how people think about their own abortion attitudes rather than the metaphors people might use to talk about abortion itself. In our analysis we extend the use of conceptual metaphors for the analysis of abortion attitudes. To our knowledge, there were no similar studies reported in the literature. [8]

2.2 From conceptual metaphors to generative metaphors

Metaphors are commonly thought of as descriptions of one thing in terms of another. LAKOFF and JOHNSON (1980) developed theoretical distinctions across various types of metaphorizing. They called one of those metaphor types conceptual which refers to the proposition that people use metaphors in their thinking—about experiences and things. If we accept that our thinking is in part structured through metaphors, empirical questions surface, such as—are there common human metaphors that work cross-culturally to structure our thinking on some common phenomena? For example, researchers have studied the conceptual metaphor "Love is a journey" across multiple cultural contexts and found it to hold up. Not all conceptual metaphors did. This and other critiques have led scholars toward both important advancements and articulations of limitations in conceptual metaphor theory. [9]

For our purposes here, we accept the idea that metaphors are involved in how we make sense of our experiences, our relations to things in the world and ourselves. We do not intend to suggest a one-to-one correspondence between a metaphor and a thought, nor do we mean to argue that metaphors are the only ways in which thought is structured. Instead, what we argue is that our understanding around the thinking involved in complex attitudes might be well-suited to analysis using metaphors. [10]

As pointed out above, researchers have primarily used metaphors to reconstruct what things might mean for participants given their use of metaphors. There have been a smaller number of studies through which researchers focused on the conceptual level of analysis. Those who did so ran into problems applying conceptual metaphors across a corpus of data—for example, when looking at a body of data, researchers struggled to find participants applying the same conceptual metaphors used at the level of specificity and systematicity that would be necessary to establish the analysis as empirically sound (GIBBS, 2011). The myriad ways participants might use metaphors to talk about their attitudes makes the task of systematization more difficult. Rather than using an analysis of conceptual metaphors, we moved toward SCHÖN's (1979) generative metaphor without letting go of the possibility that conceptual metaphors might be entailed in how people organize their complex thinking. [11]

One of the things people employ metaphors to do is to indicate a pragmatic self-understanding, including the relationship between ourselves as beings and the attitudes we are able to express (MELO-PFEIFER & CHICK, 2022; ROBERTS, 2004). In our application of this theory, we are particularly concerned with the ways metaphors help us understand how people manage the complexities of their own abortion attitudes in relation to how they understand themselves as holders of those attitudes. [12]

According to SCHÖN (1979), a metaphor is generative precisely because people can draw on it to open up possibilities for thinking in new ways. Generative metaphors are often tacit to our thinking and help to bring a set of assumptions into a coherent perspective or story. One task of social scientists might be to critically assess the generative metaphors that are implicated through peoples' conceptualizations. Reconstructing such metaphors from micro-linguistic events is an emergent process. Once articulated, the generative metaphor can reveal taken for granted aspects of one's thinking (VADEBONCOEUR & TORRES, 2003). There is a recursive relationship between recognizing and articulating the generative metaphor and further articulating it to turn it back toward the data in a more systematic way. [13]

2.3 Using generative metaphors as a qualitative data analysis tool for understanding abortion attitudes

A few researchers have used generative metaphors to examine social phenomena. None of these studies involved scholars looking specifically at abortion attitudes, but they set the stage for our approach. VADEBONCOEUR and TORRES used generative metaphor analysis to articulate the complexities of perspectives on teaching for both in-service and pre-service teachers. They found that the generative metaphors of practicing teachers troubled the taken-for-granted political discourses around theory and practice in teacher preparation. In their study, more complicated thinking around teaching was evidenced through the generative metaphors that emerged in the teachers' talk. VADEBONCOEUR and TORRES articulated individualized generative metaphors for each of their teacher-participants and then contrasted them. There were common generative metaphors across the teachers which led them to distinguish between metaphors capable of undermining or facilitating either/or binary conceptions with respect to the relationship of theory to practice. Researchers were able to distinguish complexity of thought through the complexity of the generative metaphors associated with the thinking. This is an important potential outcome for using generative metaphors as analytic tools. [14]

More recently, LOGLER, YOO and FRIEDMAN (2018) noted that it was difficult to use generative metaphors in research and so they developed a "toolkit" for using them as design-based research implements. They articulated a four-step process that included gathering rich experiences from the community, composing a set of generative metaphors, making visual metaphor cards, and then bringing those cards into research with a target demographic. They employed metaphor cards to concretize a variety of generative metaphors in particular ways that they then used in the research process to spark new ideas and new ways of thinking about the phenomena of interest with participants. On a generative metaphor card developers would include, for example, the name of the metaphor, an image of the metaphor, and lists of the characteristics of the metaphor that would be presumed relevant to the phenomena of interest. [15]

In the study of attitudes, researchers have been using the metaphorical binary "opened/closed." This binary metaphor has been largely employed without reflection, but in a way that indicates its unquestioned cultural relevance to our mainstream conceptions of attitudes (e.g., BECKER, CHASIN, CHASIN, HERZIG & ROTH, 2013; EASLER, HAUETER, ROPER, FREEBORN & DYCHES, 2018; HENEHAN & SARKEES, 2009; MASON, BOERSMA & FAULKENBERRY, 1988; NISBET, HART, MYERS & ELLITHORPE, 2013; STETS & LEIK, 1993; WHITE & GREEN, 2011). STETS and LEIK (1993) specifically talked about their study of abortion attitudes using the metaphor of opened/closed. Scholars who referred to abortion attitudes as "opened" or "closed" were generally using those metaphors to indicate the likelihood that people would change their attitude—in other words, how opened or closed a person's attitude to change is. The "doorway" metaphor helped us articulate the more subtle aspects of that opened/closed metaphor with respect to abortion attitudes. For example, if interviewees talked about being "pro-life" but being willing to understand someone choosing an abortion if they had been raped—that interviewees were indicating that their "pro-life" attitudes were open to accepting someone's abortion choice in the case of rape. They were not indicating a change in their attitude, but an openness to an other's experience and decision. However, other "pro-life" interviewees might say that in all situations abortion is wrong and their attitude might be completely closed to an alternative perspective or experience. Our use of the generative metaphor "doorway" extends the complexities with which researchers can avoid the binary oppositions of opened/closed and pro-life/pro-choice while facilitating the possibility to see new complexities and multi-dimensionalities. [16]

When the phenomenon of interest is one's attitudes, the phenomenon cannot be disentangled from one's ideas of who one is. As will be demonstrated in the findings section of the paper, we used the metaphor to analyze the complex ways in which people positioned their attitudes in relation to themselves. When people claim to hold this or that attitude, they indicate who they think they are in relation to the topic at hand. More importantly, they indicate a sense about how they might be interpreted by social others. [17]

We understand attitudes to be normative expressions that establish how one's self might be recognized through culturally-situated interpretations of espoused attitudes. Such interpretive possibilities, we argue, are engaged through generative metaphors that connect one's identities with one's attitudes and afford us all an opportunity to garner fresh insights from this new orientation and to interpret the self, in principle, as always-under-revision and available to becoming new. For these reasons, we argue that the generative metaphor "doorway" is a successful one for the study of attitudes, fostering novel opportunities to explore complex relations between identities and abortion attitudes. The overarching goal of the larger project within which we have situated this study involves finding additional ways to examine complex and multidimensional aspects of abortion attitudes. Our original use of generative metaphors for analysis is particularly well-suited for this task. [18]

3. Methods

In this paper, we report on the findings from a small subset of data and analyses that are part of an overarching, multi-method, bilingual (Spanish and English), large scale empirical research project focused on articulating the complexities and dimensionalities of U.S. abortion attitudes (JOZKOWSKI, CRAWFORD & HUNT, 2018). Our umbrella project involves several iterations of both quantitative data generated through questionnaires and qualitative data generated through one-on-one zoom-based interviews. In our analysis, we relied solely on a pilot set of interview data with both Spanish and English speakers. [19]

3.1 Participant sample

Interviewees comprised a carefully selected subset of those surveyed through online questionnaires. We used a quota-based sampling strategy (MILLER, GUIDRY, DAHMAN & THOMSON, 2020) through a non-probability Growth from Knowledge (GfK) panel in Autumn 2020 to obtain a sample of 1583 adults living in the U.S. to participate in an online survey regarding social issues including abortion attitudes1). Approximately 30% of respondents preferred to complete the questionnaire in Spanish and 70% in English. Panelists were contacted via e-mail. To obtain the smaller sample for the interviews, we used purposive sampling (CAMPBELL et al., 2020) of survey participants who indicated a willingness to participate in the interview portion of the study. We applied a maximum variation approach to obtain interviewees who reflected a range of complexity scores given their responses to the questionnaire, racial and ethnic identifiers, gender, political affiliation, age, and language groups. We intended to include interviewees who, when taken together, reflected a range of complexity scores given their responses to the questionnaires. All interviewees were residents of the U.S. See the Appendix for a list of descriptors and pseudonyms of interviewees. [20]

3.2 Methods of data collection and analysis

We analyzed twenty-four semi-structured pilot interviews (ten in Spanish and fourteen in English) conducted prior to the U.S. Supreme Court Dobbs v. Jackson Women's Health Organization decision of 20222). The interviews lasted approximately one hour and were composed of open-ended questions, including hypothetical questions such as "Imagine you are making a movie or documentary about abortion, what would that film be about?" There were eight interviewers with varying degrees of experience conducting interviews. One purpose of this pilot study was to help interviewers gain experience with this specific interview protocol. Five of those interviewers are co-authors on this paper and two of the five are Spanish/English bilingual. The pilot interview sampling procedure and protocol used in the present study were subsequently taken up in the main study with minor tweaks, but those analyses are not yet completed. [21]

We conducted semi-structured, hermeneutic interviews (CARSPECKEN, 1996) using Zoom, with participants choosing whether to turn their cameras on. Interviewers always had their cameras on. We chose this approach to interviewing because the open-ended, semi-structured nature of the approach allowed interviewees to talk freely about this culturally-sensitive topic and it allowed interviewers to ask unbiased questions that were relevant and attentive to the personal perspectives of the participants. The audio-recordings were transcribed and then each interviewer created a thick record, which included pre and post reflections. This study was approved by the Institutional Review Board and informed consent was provided prior to data collection. [22]

To analyze the data, we (the first four authors) followed steps similar, but not identical, to those outlined by LOGLER et al. (2018).

3.2.1 Establishing deep familiarity

We (the first four authors) read through the corpus of interviews multiple times. We met weekly to discuss our interpretations. Our original interest was to note episodes when interviewees expressed Aha moments with respect to their own attitudes—moments when perhaps they realized their attitudes were more complicated than they had originally expected. There were not moments in the interviews when participants changed their minds, but there were moments in the interviews when participants talked about alternative perspectives or ways of thinking about abortion and demonstrated an un/willingness to reflect on their own attitudes given those alternatives. [24]

We individually marked those moments across the interviews and engaged in meaning field analysis (CARSPECKEN, 1996; DENNIS, 2020) to notice the meaning of those alternatives for the interviewees. We used the qualitative software Dedoose to collaborative code the interviews using our meaning field analyses. We then peer debriefed our initial coding. The goal at this stage was to gather the richness of the data relevant to possible complexities in one's abortion attitudes. [25]

3.2.2 Articulating the generative metaphor

Following our initial coding, co-author Kathryn LaROCHE created a table that included each interview and key segments of the data through which interviewees 1. considered alternative perspectives and 2. reflected on their own attitudes. We noted that interviewees expressed their attitudes by taking a third person position toward them. Their expressions indicated both a claiming of attitudes and a sense of self—that is, who they were when they made those claims. Interviewers asked follow-up questions to better understand the convictions and contradictions interviewees articulated when sharing their abortion attitudes. Such follow-ups afforded opportunities for us to join with interviewees in a collaborative exploration of abortion attitude complexities including points of strength, in/flexibility, absence, movement, and un/acknowledged alternate perspectives. [26]

Through those dialogues, the generative metaphor began to emerge. The methodological practice here was similar to focused brainstorming. We selected features to pay attention to in the analysis. We let a lot of ideas surface around the patterns in the coding and we paid attention to the metaphorical words and gestures we found ourselves using in conversation. Our words and gestures began to coalesce around the contrast between openness and closedness and through the visual metaphor of a window. The goal of this stage was to select, as a team, salient features relevant to attitude complexity using the contrast we were coalescing around. Co-author Katie HAUS proposed the idea of the generative metaphor door and we reached consensus on this. [27]

3.2.3 Creating/developing the generative metaphor

We further created/developed the generative metaphor door across iterations that were aided by visual drawings produced by Katie who used a stylus and laptop to create simple black and white drawings. These drawings facilitated the development of the metaphor "doorway" by allowing us to visualize how doors opened and closed and how they worked more subtly. We created a list of components of doors. We represented a participant's attitude toward abortion, for example, support for abortion as the door itself. The drawings became more complex as we developed the varying aspects of doors and doorways. Each part (for example, hinges) was drawn and labeled. In each label, Katie included broad descriptions about how the component functioned within the generative metaphor. For example, doors that opened might cross over a threshold. Hinges functioned as moving parts to open and close doors. We articulated new aspects of the metaphor, spoke back to existing components, and considered possible ways this metaphor had yet to be articulated. After each discussion, in returning to the data, we continued the conversation while reading through each transcript and considering these things for ourselves, saving them to bring back to our weekly meetings. As we identified new elements of the metaphor, Katie added new drawings and refined the labels on the original drawings to better reflect our emergent understanding. The drawings facilitated a more detailed development of the generative metaphor. The goal of this stage was to fully constitute the metaphor and its features. Through this dialogic process we refined the metaphor "door" to "doorway" because it created the space to locate the personal identities of participants in relation to their doors (attitudes). Their doors were composed of moving and non-moving elements. Through the simplicity of the drawings, we were able to examine the structure of participant attitudes while acknowledging that the meaning of participants' ideas went beyond the limits of the drawings, and beyond the textually described labels and ideas of the metaphor itself. [28]

3.2.4 Applying the generative metaphor

Once we had a list of the components, we returned to the data. We added to the original chart created by Kathryn to locate each interviewee's perspectives according to our metaphor. The door and doorframe were used for every participant. In this way, we were careful to always identify interviewees' orienting abortion attitudes as their "doors" and to note what interviewees based their attitudes on as their doorframes. After that, the component elements of the doorway were identified as relevant to any one particular interviewee's way of talking about abortion attitudes. For example, for some people there was a doorstopper, something external to them or their own attitudes that kept their door from opening wider to other perspectives or alternative attitudes. For other participants, their doors had locks. We returned to each interview, locating the elements and adding elements as needed. By mapping interviewees' talk about their abortion attitudes onto the imagined structures of our metaphor, our interpretations better articulated the complexities that were connected with people's openness to alternative attitudes and perspectives. In revisiting, revising, and interrogating the metaphor through ongoing use, we were able to understand the ways a bounded metaphor restricted and supported our analyses and we were able to keep the analyses closely tethered to the data. [29]

3.3 Validity

ARMSTRONG, DAVIS and PAULSON (2011) demonstrated that multiple analytic methods can be used to validate a metaphor analysis. We began with/returned to other reconstructive approaches to analysis to anchor our interpretive validity (particularly critical analyses of meaning fields and reconstructive horizon analysis following CARSPECKEN, 1996; DENNIS, 2020). Returning to the data with the generative metaphor as an analytic tool carried with it some interpretive risks (SCHMITT, 2000, 2005). First, we needed to be sure we were not force fitting the metaphor onto the data. We checked this by looking for counter examples, peer debriefing, multiple readings of the data, and refining the metaphor. This metaphor did not explain or account for everything worth analyzing in the data and was not intended to do so. Second, we needed to be transparent about how we interacted with the data via the metaphor. In this way, we situate the validity of the inferences in public view. Third, we needed to identify the limits of the metaphor. We acknowledge that this metaphor, like any metaphor, will inevitably fall short of being able to fully explicate all the possible complexities in abortion attitudes. [30]

4. How the Metaphor Worked Analytically

Our illustration of the "doorway" metaphor includes a detailed description of the components that emerged: the frame, the door, the doorknob, hinges, doorstoppers, and drafts that blow in between the threshold and the door. The descriptions of these components constitute the first part of this section of the paper (Section 4.1). We follow it with an exemplar—that is, an interpretation of one particular interviewee's attitudes on abortion as reimagined through the metaphor (Section 4.2). The door itself reflects the participant's identified attitude toward abortion. Each of the components works with the door. Every attitude had a frame, but other components were not always present. Taken together, the metaphorical components indicated how this interviewee structured her attitudes in relation to others' experiences/perspectives. [31]

4.1 The components—A detailed description

Figure 1: The door [32]

4.1.1 The frame

Imagine that the door is the particular abortion attitude. Whatever that particular abortion attitude was, it was framed through a certain conception. We defined the frame as a non-moving part of the doorway and it holds the door (the attitude) and structures the doorway. We interpreted interviewees' universalizing claims or ideas about abortion as their doorframe, indicating relative stability in their abortion attitudes. Doorframes supported and shaped participants' abortion attitudes, as interpreted through the doorway metaphor. Common doorframes included religiosity, family influence, ideas of life, rights, and autonomy. For example, religion functioned as a frame for Susan, a 61-year- old Black/African American3) woman who espoused an anti-abortion attitude. She said "I believe in Christ; he gives us life. So it's formed, I believe in that, you are killing." [33]

Religion was not the only doorframe. Sara, a 22-year-old Latina reported

"I don't agree [with abortion]. I don't see it as something viable, but the option should be there ... in situations I mentioned ... it should be used, I don't know, responsibly and not just to escape, I don't know, to escape from the past." [34]

Her doorframe was her family upbringing. She repeatedly mentioned that her upbringing still framed her abortion attitudes, even though she was now more open to the complexities of abortion decision-making. How she was raised was framing her attitudes even when they changed. In this example, family upbringing served as a doorframe. Sara noted:

"Yes, well, like any child, in your family, I think, they lead you down a certain path, they teach you certain things that as a child you don't even realize there is another perspective, any other perspective that is different from your family and well, for example, my house is a Christian family, the abortion issue, pagan issues, do you understand? Well, like I said, it's like a little while ago, like you have an idea, it's something predetermined, it's something you were taught and everything is like this like they taught you and when you grow up you realize things aren't like this, that there is something beyond what they taught you, right? So I, for example, in the church, my family, taught me this is how it would be; I'm going to put it in my own words, I'm not going to say exactly what they said. But, for example, you heard that it is like you killed someone, that it is a baby, a living being, that you can't kill someone, that is a sin. But now, now that I have my own thoughts, let's say, no one is telling me, 'You have to think about it like this.' I would think, let's see, it's very complicated. On one hand, I haven't removed this [upbringing frame] from my ideas, that I am killing someone, that it is a living being, because in the end, it is, right? It is a baby that is growing in your womb, but on the other hand, there is the situation I described before, I don't know, of a rape or, for example, there are other situations like when the child is sick, later its life will be chaos in terms of health issues, right? So it's like, for example, I would agree with them saying it [abortion] should not be accepted." [35]

When asked about people who identified as both pro-choice and pro-life, Mark (middle-aged, white man) told his interviewer:

 "Yes. I mean, I kind of feel that I'm that way. As far as on a personal level, I feel like I would be pro-life in my [own] decision making but again I'm also at a stage in life where we're financially sound, we've got a family. Yeah. So, on a personal level, I'd say I'm pro-life for myself and my family, but for you and your family, I would say I'm pro-choice because that's your decision for you and your family to make, not me." [36]

Others held similar sentiments, noting that their perspectives toward abortion were intended to apply only to themselves. Mark had established that for him, at this point in his life, he did not believe he would choose an abortion. He called this "pro-life." However, he knew he was choosing based on his own circumstances and he believed everyone else should be able to choose as well. Here it seems that "choice" framed his current "pro-life" door/attitude. [37]

Already, with only two components articulated, we recognized that the metaphor helped us to articulate and elaborate complexities across the ways in which various commonly studied factors, such as religion, function in relation to people's own way of thinking about their abortion attitudes. We drew on an array of additional possible parts to continue the analysis and further our articulation of the myriad ways people positioned themselves in relation to their abortion attitudes. [38]

4.1.2 A window

The presence of a window in a door symbolized individuals' ability to think beyond their own frame and acknowledge the existence of other perspectives and ideas. Remembering that the door itself represented the interviewee's attitudes toward abortion, a window was a way interviewees expressed an ability to peer out from that attitude toward someone/some perspective or experience that they cast as outside their own attitude and its frame. A window demonstrated that interviewees were able to understand their own positionality in relation to attitudes or practices held by others. In the following example, Ruth, a 45-year-old Latina, who espoused an anti-abortion attitude (her door), acknowledged that other perspectives existed, particularly thinking through this with respect to her daughter. She discussed a hypothetical scenario in which her daughter considered an abortion:

"The truth is that everyone has to make their own decisions, the only thing I can do is advise her, right? I can advise her, and I think I would respect it, I think I would respect it, but I think I would tell her that we would never, ever address the issue again, because she knows that ... if she did it, I think it would stay there. And we would never, ever talk about it again, but I would respect her decision. The truth is that everyone has to live their own life." [39]

It was as if she could use a window blind to intentionally unsee something through the window. The window allowed her to acknowledge that everyone (including her daughter) has to live the own life, BUT she herself was opposed to abortion and unwilling to even talk about it (even with her daughter). [40]

4.1.3 Doorknobs

Imagine that for some interviewees, there was a doorknob on the door which made it possible for people to imagine their door being opened—that is, the interviewees situated their attitude as capable of being opened to ideas, perspectives, practices, or decisions that were outside their own particular frame. We understood doorknobs to exist when participants reflected on their acknowledgment of alternative perspectives—that is a person's doorknob was interpreted as locked when he/she expressed not being "open" to alternative perspectives, not letting those perspectives affect own attitudes. Where windows were understood when participants acknowledged alternate perspectives, doorknobs became evident through further exploration. Doorknobs were interpreted as unlocked when an interviewee was able to consider alternative perspectives. Referring back to Ruth (quoted above), she indicated that her daughter's decision in the hypothetical situation would not make her reconsider her perspective, indicating that her own belief at the time was not open for change or insight even given hypothetical actions on her daughter's part. We interpreted Ruth's window to be the difference between her perspective on abortion and her willingness to imagine her daughter espousing a different perspective. The locked doorknob represents her refusal to reconsider her own attitude and its framing. This example illustrates how the parts of the doorway function in relation to one another. [41]

Another participant, Mary, a 71-year-old white woman, was a former nurse who considered abortion a medical decision. This attitude (or door in our metaphor) was framed through the concept of autonomy. Mary said she would support people regardless of their pregnancy decision so long as they were the ones who made them. She said that if she were to produce a documentary on abortion, she would include

"... the different reasons why someone would have an abortion and it might be rape and they don't want to carry the baby, while some women might say, I want to carry the baby and raise the baby or have the baby adopted. And that's based on their own beliefs. So, there's no right or wrong in that regard." [42]

Her Catholic belief in the sanctity of life functioned as an unlocked doorknob because it compelled her to consider both the lives of the unborn and the autonomous lives of pregnant people. Her attitudinal door (abortion as a medical decision) was unlocked, as she reflected on the different perspectives and situations that might inform and influence her attitudes toward whether abortion was the right choice for any particular individual when they went through the decision process. These sanctity of life considerations were not merely medical decisions which is what she had identified as her attitude toward abortion. [43]

In our doorway metaphor, the doorknob was interpreted as a moving part of the doorway, whereas a window was not. Thus, aspects of one's attitude that are understood as doorknobs are aspects that interviewees can move and engage with—use to open their door (attitudes) from within the frame, whereas windows are non-moving parts that one looks through—passive and visual. [44]

4.1.4 Keys

In some instances, doors were locked while in others, participants identified keys that might unlock their doors. A key was inferred to exist when participants noted conditions or reasons that could sway their own perspectives. While unlocked doors were more flexible in terms of the interviewee's openness toward different perspectives, doors with keys had highly specific hypotheticals that were required for participants' stance to recognize, appreciate, relate with, or otherwise engage with perspectives that were different from their own. "Keys" often included specific scenarios like rape, incest, or the health of the pregnant person and/or fetus. [45]

When asked about a scenario where abortion would be the best option, Andrea, a 38-year-old Latina, noted "Well, when she's in danger of dying. When she has cancer, when she has AIDS, when she is on the verge of death. When there are diseases that put both at risk, it would be a good option." These scenarios worked as the key for Andrea's door, which developed in her description of abortion as "irresponsible" as there was "no excuse [for having an unplanned pregnancy] ... it is really a negligence on the part of the woman." Not all locked doors had keys, as some participants did not identify any scenarios that would influence a change in their perspectives or create an openness to understanding an alternative point of view. [46]

4.1.5 Hinges

Hinges between a participant's door and doorframe were evident when participants considered certain things on a spectrum, much like the oscillation of a door hinge, such as the timing of a pregnancy, the variety of reasons identified for why someone may consider an abortion, or the ideas participants had about potential violence or bad relationships between pregnant women and their partner. For example, when elaborating on his belief that abortion is wrong, George, a 58-year-old white man, stated: "In the early stages [of a pregnancy], I could accept somebody doing that [having an abortion]." For him, his anti-abortion attitude was open to abortion in the early gestational stages. This hinge swings his attitudes open to other perspectives and actions that he claimed he would not want his female partners to take. [47]

4.1.6 Doorstoppers

Figure 2: Doorstopper [48]

Some participants did not have factors like those noted above in their perspectives, suggesting doors that were locked or hinges were rusted shut. In further analyzing components of flexibility (e.g., hinges or doorknobs), some interviewees implied sticking points, which we understood as doorstoppers. Doorstoppers were only functionally recognized if there were working hinges. These doorstoppers halted how wide the door would swing open. Doorstoppers were fixed components through which participants reinforced their own perspectives. In other words, their attitudes might swing open, but eventually hit a stopping point and opened no further. Doorstoppers indicated limits in their own ability/willingness to consider different perspectives. [49]

Sometimes the doorstoppers reflect factors within people's own identities, such as a personal experience with parenting. Others' doorstoppers marked a lack of personal experience. Still others reflected a staunch abrupt halt in considering alternative perspectives as going "too far." For example, Emma, a 41-year-old Latina, who was opposed to abortion, reflected upon a friend's abortion disclosure, and said: "So I think I can't understand because I've never had an abortion, and women who have had an abortion, I think we have no idea about the impact that an abortion can have psychologically as well." Having not experienced an abortion herself, the psychological impacts her friend reported to her functioned as a doorstopper in opening the door of abortion support more broadly. Emma's door was "pro-life" framed through her Catholic religion. However, her compassion functioned as a hinge. For example, she expressed concern for the situations of both rape and unplanned young pregnancies (where the root problem is not solvable through abortion). Emma's friend had an abortion and shared the negative consequences of that abortion. While her door opened through compassion, it met a stopper—that there were more negative consequences to abortion and one could not know these if one had not known how damaging an abortion could be without having had the experience. Emma's friend's negative psychological experiences served as a doorstopper to her opening up further to the situations of women for whom she might find compassion. [50]

George's abortion attitude had two sides—he believed it was morally wrong but should be legal. His doorstopper was gestational age—early on he could accept abortion being legal and possibly moral. Though he thought abortion was morally wrong and legally necessary, he was open on both aspects most specifically in the early stages of gestation. Past the early months, George did not support the legal right to an abortion. [51]

4.1.7 Backdoors and thresholds

In some instances, participants seemed to have stepped through one door in the past before coming to their present perspective. We interpreted this as participants having a back door. In articulating a past change in perspective, Marta, a 25-year-old Latina, claimed:

"Yes, I believe that when I was younger, when I didn't have children, my opinion was pro-life, pro-life ... But now that I have children and I see what it means to take care of children and I put myself in the shoes of people who are not called to be mothers, who do not have the resources to raise children and even more those who were maybe abused, raped, my opinion changed a lot." [52]

Another interviewee talked about previously supporting a woman's right to choose an abortion, but after becoming a parent, it was impossible for him to imagine an abortion without thinking about his kids being aborted. For others, it was a reminder of important components of their own attitudes, or underlying beliefs about people who pursue abortions. Other participants also linked their own age and maturity with changes in their perspectives, which we considered as factors reflecting a metaphorical threshold. [53]

In stepping over the threshold of that previous doorway by way of a particular or collection of moments, events, and experiences some people reported that they came to new perspectives. While they acknowledged the way this old door was different from their current door, it still existed behind them. Mark, a 47-year-old white man, described several factors that contributed to this change in his perspective, noting:

"So, I'd say once I kind of started to give these things thought, a teenager getting into being promiscuous and things like that, I was very much, 'I'm a guy, it's not my choice.' And I think that's actually kind of a cop-out truthfully ... when I was in college, I had a friend whose roommate in a single year had three, and I'm like, okay, you're using abortion as a form of birth control essentially ... I know for myself, I've got three children and I was much looser about abortion pre-children." [54]

These markers of maturation, personal experience, reflection, and parenthood resulted in a new perspective where "... now as a parent and being married, if my wife, if we were to have an oopsy, abortion wouldn't even be on the table for us." Having stepped through the threshold at the back door and into their current doorway, this participant referenced these factors as influential in the shift toward his current perspective. [55]

4.1.8 Complexities—Drafty doors and relationships across the components

Figure 3: A drafty door [56]

Some participants also indicated gaps in their perspectives where the presence of conflicting ideas was evident. We understood such gaps through the image of a drafty door, where a gap between the threshold and the door itself allows gusts of wind to blow through. Such gaps were not always acknowledged by participants. Drafts indicated that conflicting or alternative thoughts could blow into their mindset when thinking about their abortion attitudes, without indicating a particular openness to those attitudes. In contrast with windows, drafts actually stirred up a way of thinking about abortion attitudes as they entered the doorway whereas windows just provided a way to see what others' alternative perspectives looked like without those perspectives churning up one's own attitudes. Sandra, a 46-year-old Latina, described her abortion attitudes as a personal moral decision, framed both by a study/belief in the Bible and by the liberty to decide what is right or wrong for herself. She stated:

"When I studied the Bible, that's how I grew, in my opinion of abortion, whether it's right or wrong... If somebody asks me, I'm not going to shy away, and say that I think it's wrong. Because I should have ... I mean, you have the right to do what you want to do. I have the right to believe the way I do, too. I'm not going to stop you. I'm not going to fight the laws. I'm not going to do all of that. But when it comes to a personal decision, that's my personal belief and decision, and I'm not here to judge. I'm just here to do what I think is right for me. That to me is really part of my faith Is living by what I believe the Bible teaches. That's where I stand." [57]

This participant believed abortion was wrong, which was informed by her reading of the Bible. While her door was locked, as she stated "... I don't think that it's [abortion] okay under any circumstance," she was also reluctant to label other people's actions as wrong, to limit those actions, or consider limiting people's right to do as they chose due to the messaging of the Bible, and her consideration of choice as a right. While she believed abortion was unacceptable in all circumstances (representing no keys), she did not want to limit people from making their own decisions, even decisions she would disagree with—hence the dual frame (Bible and the liberty to decide for oneself what was right or wrong). This dual framing left a blustery gap beneath her uneasy knowledge that people would still make decisions she disagreed with and believed were morally wrong. The subtlety here is interesting because Sandra located her attitudes toward abortion through a religious frame—but it is a personal religious frame. She conceptualized abortion as a personal decision which for her is framed by her reading of the Bible. She distinguished her own attitude toward abortion from the law and from a judgement about others' attitudes or decisions. She felt the presence of both (as a draft), but they had nothing to do with her own frame or the opening of her own door. [58]

The doorway metaphor was a detailed, subtle, and structural way for us to grasp how participants were positioning themselves in relation to their own attitudes. The various components provided us with ways to articulate some of the complexities we had heard in the interviews. The metaphor helped us:

Though all participants' perspective were unique, we were able to dually understand each doorway to be individual, comprising the components they identified during the interviews and yet, simultaneously note that the functions of the components were similarly structured and systematically applied across participants. This addresses a limitation in the field of using metaphors to analyze qualitative data (GIBBS, 2011) Below we demonstrate how we conducted the metaphor analysis by illustrating its outcomes for one participant. In a future paper, reporting on the substantive findings of the analysis, we illustrate how the generative metaphor as a tool functioned to locate complexities and multi-dimensionalities across participants in ways that the more typical metaphor analysis would not be able to achieve. [60]

4.2 Doorway: An example of the generative metaphor as an analytic tool

In this section of the paper, we specifically apply the generative metaphor to one interview, and show how the metaphor worked as an analytic tool in the study of abortion attitudes. Remembering that some of the components of the door are fixed and some are moveable, we describe the door, the frame, and various relevant components for one of our interviewees.

Figure 4: Nancy's door [61]

We reconstructed Nancy's (a 49-year-old Black/African American woman) doorway because it was especially complicated. Nancy's door, her attitude, was liberally rights-based and health-based. She believed that abortion was always a woman's right in all situations. From a social viewpoint, she said, "I think of women's rights, first and foremost." She also described abortion as a health issue, noting that "it's necessary for women's health and mental health." Asked to consider when she would feel uncomfortable with abortion, similarly to what we discuss above, she stated that "The last trimester ... makes me rather uncomfortable, but I wouldn't necessarily exclude it [from being legal]." As the fetus developed more during this stage, she felt less comfortable, yet she also referenced the possibility that given specific scenarios she could still justify her support for the continued availability of abortion even at this later stage in the pregnancy. While her feelings of discomfort with third trimester abortion seem to conflict with her beliefs that abortion should be legal and easily accessible, she mentioned that her discomfort should not mean abortion should be restricted in these situations. [62]

We interpreted an active sense of agency on her part in dealing with her own subjective sense of discomfort and its contradiction with her stated attitudes of women's right/choice and health issue. She acknowledged that these feelings were her own—they were not being inflicted upon her from outside forces. She felt that the wellbeing of the pregnant person and potential pregnancy should always outweigh others' discomfort, including her own. For this reason, we thought her prioritization of others' health decisions over her own feelings and awareness of discomfort were part of her door, like one made from multiple pieces of wood or metal, perhaps with parts manufactured in different batches before being fused into a final product. She actively integrated her own discomfort, a subjective feeling state, into her perspective that abortion was always a woman's choice and always a health issue (normative and evaluative claims about what she considered right). In other words, she distinguished her own (or others') discomfort from what she believed should inform the law or access to abortion as a necessary medical procedure. The right to abortion access for those who needed it was central—she considered myriad women whose experiences and needs may have been different than her own, choosing to center them instead on her own internal experience. [63]

Nancy believed abortion was a woman's right and a health issue. What supported or held those beliefs in place was a rich composite of social beliefs. Early in the interview, Nancy described the way her family shaped her attitudes, stating, "... I was raised in a very liberal family. My grandmother's case ... was actually the landmark decision that led to the Supreme Court decision on equal rights for women." She described how her family history directly led to her own initiative as an activist, where she accumulated experiences supporting her friends. She noted "I was the girl who went with all of my friends ... [when they had abortions]." Coinciding with this family context, Nancy developed a description of family and family roles (including the labor of child-rearing) as social configurations that were not dependent upon blood or biological linkages. She said, "I didn't have any biological children, but I married a man that had children himself and I have a 12-year-old granddaughter and a five-year-old grandson." Her atheism was also a feature of her frame: "... I personally am an atheist ... I don't think that religion has any place in medical decisions whatsoever." She used this distinction to support her ideas that abortion is medical care. She also referenced ideas of population control in her framing, stating "We have enough damn people on the planet already. We don't need more people. We simply don't. I mean, the environment, this pandemic, I mean, the planet is telling us to get the hell off." This social perspective framed her attitude that abortion was a rights-based, and health/medical-based decisions/choices. [64]

In her door, there was a window. It was through this window that she was able to explain others' attitudes toward abortion, without agreeing with those perspectives. When asked to describe whether she could understand others who are not supportive of abortion, Nancy spoke of her aunt's marriage to a Buddhist woman:

"My aunt married a woman...and she's Buddhist...and she clung to her faith. And when people ask how old people are in that culture, they always add nine months because they believe that life begins at conception. So, a three-month-old baby is a year old in that system." [65]

She also described tension with what she called fundamentalist Christians, noting "I'm empathetic and I'm respectful to a point of their beliefs, but ... I don't believe what you believe. And you can't force me to do so." Her acknowledgment of beliefs different than her own functioned as a window into different perspectives, but these alternative perspectives were situated through her door and frame, without affecting them. What she described through the window was not something that would cause her door to open. [66]

Relevant to her frame, Nancy articulated a threshold. She stated that

"[t]he only thing that I can absolutely think of [that would change my perspective] is if they were to find some way to prove that the essence of what makes us "us" is imparted the moment that the sperm fertilizes the egg." [67]

Nancy also later mentioned that she believed "it's that first breath that is I'm taking the energy in that is going to be me." For her to no longer believe abortion is acceptable in general, this is the knowledge she believed she would have to accumulate. We might label her threshold as the essence of when a being is scientifically established—currently she thought science says "first breath" is life, but if that changed then she could imagine her attitudes changing. While the threshold itself is not moveable, her position could shift if she passes through that threshold and onto a new "door" or attitude.

Figure 5: Nancy's threshold [68]

The door, frame, window, and threshold all describe non-moving aspects of how she positioned her attitudes toward abortion. Next, we'll look at components related with the potential for movement in her attitudes. The hinges of her door (connecting the door with the frame, but also providing movement—opening her door/her attitude toward abortion) were constituted of concerns for third-trimester abortions and the visualization of the fetus as looking more human, having human features.

Figure 6: Nancy's hinge [69]

There was a hinge on Nancy's doorway that functioned as a moving part, making it possible for her door to open to other attitudes and perspectives. Nancy mentioned several times that she was uncomfortable with third-trimester abortion, even saying "... I would say I would be more comfortable with a ban on everything post third trimester." Here, we noted her feelings toward abortion hinging on this third trimester. Understanding that she was actively mediating her own sense of discomfort as relevant to her abortion attitude and recognizing her social frame, this hinge connected that active mediation of her discomfort of third trimester abortions with her human social concern and justice. We identified the timing of abortion as a hinge in Nancy's case, as the timing of abortion seemed crucial in her consideration of limiting abortion—a perspective she identified as different from her own. She was unwilling to open the door to accepting abortion restrictions on the basis of her own discomfort, but was, however, willing to open the door to supporting abortion restrictions based on an idea that a third trimester's fetal characteristics were so human-like.

Figure 7: Nancy's doorstopper [70]

Nancy explored the tension between this hinge (third trimester abortion) as a connection between her attitude (door) and what framed that attitude. In so doing, we identified an interpretive doorstopper. The hinge indicates movement or possible opening based on the third trimester characteristics of the fetus. This hinge could swing Nancy's attitude open to the idea of restricting abortion, but the door stopper reduced how far open that door swung toward the alternative perspective that perhaps women should not always have the right to choose. Nancy's doorstopper was a lesser-used and ironic interpretation of the Bible. Let's look more closely. Nancy said that

"the last trimester, when there's a lot of activity and it actually looks like a human baby, I'm a little bit more uncomfortable with that. But like I said if we're using Biblical justification, it was perfectly fine after that baby came out to kill it to up to a year old." [71]

Nancy was adamantly opposed to fundamentalist Christian "pro-life" views which she considered to be more "pro-birth" than "pro-life" and also not aligned with her interpretation of the Bible. Nancy ironically invoked the Bible as an authority to support a view opposite to that of fundamentalists who invoke the Bible (in selective ways according to Nancy) to support limiting access to abortion. She used the Bible to stop herself from opening the door too far in the direction of limiting abortions. Nancy stopped herself from continuing to explore bans on abortion given the timing hinge through which the door had originally opened. By positioning herself in relation to a counter perspective she strongly disagreed with, she limited the way she talked about abortion and tacitly created a barrier/stopper in exploring potential legal restrictions on abortion, an idea that directly contradicted her belief that abortion should be legal regardless of her feelings. Imagine the door is opening up through the hinge of fetal development/being "more human," but the door on this hinge does not swing wide open or come loose. Instead, the door swings past the threshold we called "the essence of being" and stops when it hits the door stopper of a counterhegemonic Biblical notion of permissible infanticide (Bible is seen as outside her door frame). She sets it up as a stop to her own opening notions/discomforts around the third trimester and fetal characteristics of humanness. [72]

We demonstrated how aspects of Nancy's attitudes function in relation to one another as she positioned her own views on abortion through this illustrative example. Readers can see that the complexity of Nancy's attitudes is not oriented toward change in that attitude, but toward the componential structure of her abortion attitudes and her relation to those attitudes, particularly contrasting her views that third trimester abortions involve fetuses that are active and look human. [73]

5. Discussion and Conclusion

Many Americans know that talk about abortion in the U.S. is characterized in the popular media as polarized and divisive (SAAD, 2020). Through that discursive characterization, people often implicitly hide complexities in attitudes and in their position toward their own attitudes—that is, whether they think of their own attitudes as being opened, fixed, determined, closed, coherent or unproblematic, and so forth (COHEN et al., 2007). [74]

In this paper we introduced the utility of a generative metaphor as a tool for analyzing complexities in attitudes. The model created from the metaphor "doorway" is constructive as a heuristic and didactic tool for describing how various aspects of one's attitude function in relation to one another from the perspectives of participants themselves. We found it particularly valuable for illuminating morally complex and controversial issues about which people are likely to hold attitudes. We applied the generative metaphor across two levels: as a tool for better understanding the structure of individual participants' abortion attitudes and as a tool for locating functional patterns across interviewees. In this paper we reported on the first of those. By utilizing the model independently with each person's interview, we were able to fit the metaphor within a hermeneutic understanding of that person's attitudes so that, for example, an understanding of what people's identification with religion might mean for how they conceptualized their own abortion attitudes. Then, when we looked across the analyses of individual interviewees, we found, for example, that people attributed religion in different ways—that is, religion functioned differently in how people held their abortion attitudes. [75]

Researchers, as we exemplified in our study, can employ generative metaphors to analyze participant talk on sensitive subjects especially because concepts, feelings, or knowledge related to the sensitivities may be difficult to explicate verbally (KARA, 2015; TRACY & MALVINI REDDEN, 2015). The research literature regarding abortion attitudes reflects a limitation: scholars have predominantly examined sociodemographic factors that correlate to people's perceived attitudes (for review see ADAMCZYK, KIM & DILLON, 2020), which may frequently be ascribed to the labels of pro-choice and pro-life by researchers instead of by participants themselves (BRUCE, 2020; LaROCHE, JOZKOWSKI, CRAWFORD & JACKSON, in press). Instead, we understood attitudes to be transformable, contextual, dynamic, and related with how one is identifying in terms of abortion labels, though not synonymous with such identities. As SCHÖN (1979) argued, generative metaphors offer opportunities to "see" common experiences and taken-for-granted assumptions in a new way. Results of our study suggest some new points for the study of abortion attitudes and for exploring other similarly polarized and contentious issues. This analysis affords researchers with an opportunity to imagine the potential points of dialogue and synergy across seemingly oppositional perspectives [76]

For example, in the analysis of Nancy's abortion attitudes, we ascertained that she was open to dialogue around third-trimester abortion bans even though she strongly espoused a woman's rights/choice and health attitude toward abortion. She was not open to people using the Bible as moral authority on abortion. Her articulation of points of flexibility and fixedness, with reference to others' perspectives, provided insight into perceived attitudinal openness that might create opportunities for her to reflect on her own beliefs and engage with others. Nancy's brief recognition (during the interview) of her own discomfort indicated important factors that could have provided further opportunity for her to reflect on her own attitudes. In using the Bible as a doorstopper, Nancy referenced existing perspectives she disagreed with to reinforce her stance that people's feelings and beliefs should have nothing to do with the legality of abortion. In identifying her threshold for changing her stance toward legal restriction of abortion as the point at which life begins, it is possible that her consideration of different perceived truths of when life begins, and different conceptualizations of scientific knowledge, could have led to fruitful discussion and self-examination. [77]

As exemplified through an analysis of Nancy's interview, the use of metaphor for other participants helped us to understand unspoken potential for transformation and re-evaluation of abortion attitudes (KONECKI, 2019). While Nancy espoused a strong attitude indicating an understanding of or support toward abortion as a woman's choice/right, the analysis allowed us to see more than a polarized description of her attitude would afford. We found that the newness that was enabled through the analysis (SCHÖN, 1979) involves self-recognition of her own felt tension between an espoused unwavering rights orientation and the image of something that looks human being aborted. The more the fetus becomes more visibly human in appearance, the less comfortable Nancy is with aborting it. This visual does not have a counterpart in her rights-based position in favor of abortion. We know that Nancy expressed strong pro-choice views, but those views were nuanced in ways that would have been missed through a more strongly dichotomous orientation toward analysis. [78]

While there were benefits in using generative metaphors as analytic tools in the study of attitudes, there were also limitations. We focused our use of the doorway metaphor on the functional attributes of meaning in a person's attitudes. It is important to acknowledge that this does not create a full understanding of an attitude. Careful and systematic development and application of the metaphor to an analysis of participants' orientations toward their own attitudes required collaboration and dialogue. While one might be able to test out own generative metaphors with participants (VADEBONCOEUR & TORRES, 2003), we used one-time interviews and relied on an ongoing conversation, negative case analysis, and peer-debriefing approach to collaboration to secure the validity of the model. [79]

In the study of abortion attitudes, our approach is unique. By generating a metaphor that is complex enough to depict how a way in which thorny social issues, such as abortion, are expressed in the form of attitudes we can develop new possibilities for dialogue across differences and for undoing the polarized conceptions of difficult issues. Thorough and methodical application of the metaphor raises patterns that were missed through a thematic analysis of the same interviews. [80]


This work was funded by a grant from a confidential foundation. The funder played no part in the study design, the collection, analysis, or interpretation of data, the writing of the article, or in the decision to submit it for publication.

Appendix: Interviewee Pseudonym and Demographic Characteristics


Demographic characteristics


55-year-old Latina


38-year-old Latina


52-year-old Latino


36-year-old Latino


36-year-old Latino


20-year-old Latino


75-year-old Latino


61-year-old Black woman


41-year-old Latina


51-year-old Black man


47-year-old white man


35-year-old Asian man


51-year-old Latina


65-year-old Black woman


45-year-old Latino


47-year-old white man


25-year-old Latina


71-year-old white woman


47-year-old Black man


49-year-old Black women


44-year-old Latino


45-year-old Latina


35-year-old Latino


46-year-old Latina


22-year-old Latina


1) Our quota-based sampling involved the following distributions: gender (49.5% women; 49.5% men; 1% other), age (12.8% 18-24; 17.7% 35-34; 16.7% 35-44; 17.7% 45-54; 16.4% 55-64; and 18.8% 65 and up), race/ethnicity (25% Black/African American; 25% White; 25% Latinx; 25% other), and education (30% GED, high school or less; 30% some college or associate’s degree; 30% bachelor’s degree or more). <back>

2) This U.S. Supreme Court decision overturned the constitutional right to abortion that had been nationally guaranteed through the Roe v Wade decision of 1973. <back>

3) All names are pseudonyms to protect the confidentiality of the interviewees. <back>


Adamczyk, Amy; Kim, Chunrye & Dillon, Leevia (2020). Examining public opinion about abortion: A mixed‐methods systematic review of research over the last 15 years. Sociological Inquiry, 90(4), 920-954.

Armstrong, Sonya; Davis, Hope Smith & Paulson, Eric J. (2011). The subjectivity problem: Improving triangulation approaches in metaphor analysis studies. International Journal of Qualitative Methods, 10(2), 151-163, https://journals.sagepub.com/doi/full/10.1177/160940691101000204 [Accessed: November 27, 2023].

Becker, Carol; Chasin, Laura; Chasin, Richard; Herzig, Margaret & Roth, SallyAnn (2013). From stuck debate to new conversation on controversial issues: A report from the Public Conversations Project. In Kathy Weingarten (Ed.), Cultural resistance: Challenging beliefs about men, women, and therapy (pp.143-163). London: Routledge.

Bruce, Tricia C. (2020). How Americans understand abortion: A comprehensive interview study of abortion attitudes in the U.S. Religions, 11(9), 475-480, https://news.nd.edu/assets/395804/how_americans_understand_abortion_final_7_15_20.pdf [Accessed: November 9, 2023].

Brugman, Britt; Burgers, Christian & Vis, Barbara (2019). Metaphorical framing in political discourse through words vs. concepts: A meta-analysis. Language and Cognition, 11(1), 41-65.

Cameron, Lynn (2003). Metaphor in educational discourse. London: Continuum.

Campbell, Steve; Greenwood, Melanie; Prior, Sarah; Shearer, Tonielle; Walkem, Kerrie; Young, Sarah; Bywaters, Danielle & Walker, Kim (2020). Purposive sampling: Complex or simple? Research case examples. Journal of Research in Nursing, 25(8), 652-661.

Carspecken, Phil F. (1996). Critical ethnography in educational research: A theoretical and practical guide. London: Routledge.

Cohen, Geoffrey; Sherman, David; Bastardi, Anthony; Hsu, Lillian; McGoey, Michelle & Ross, Lee (2007). Bridging the partisan divide: Self-affirmation reduces ideological closed-mindedness and inflexibility in negotiation. Journal of Personality and Social Psychology, 93(3), 415-430.

Dennis, Barbara (2020). Walking with strangers: Critical ethnography and educational promise. New York, NY: Peter Lang Publishers.

Easler, James K.; Haueter, Helena; Roper, Susanne; Freeborn, Donna, & Dyches, Tina (2018). Reasons for open and closed attitudes regarding type 1 diabetes. Diabetes Spectrum, 31(1), 37-46.

Furunes, Trude & Mykletun, Reidar (2007). Why diversity management fails: Metaphor analyses unveil manager attitudes. International Journal of Hospitality Management, 26(4), 974-990.

Garrett, Peter (2010). Attitudes to language. Cambridge: Cambridge University Press.

Gibbs, Raymond Jr. (2011). Evaluating conceptual metaphor theory. Discourse Processes, 48(8), 529-562

Henehan, Marie & Sarkees, Meredith (2009). Open doors and closed ceilings: Gender-based patterns and attitudes in the International Studies Association. International Studies Perspectives, 10(4), 428-446.

Heracleous, Loizos & Jacobs, Claus D. (2008). Crafting strategy: The role of embodied metaphors. Long Range Planning, 42(3), 309-325.

Jozkowski, Kristen N.; Crawford, Brandon L. & Hunt, Megan E. (2018). An examination of the complexity of abortion opinions in the United States: Results from Two Studies. Sexuality Research & Social Policy, 4

Kara, Helen (2015). Creative research methods in the social sciences: A practical guide. Bristol: Policy Press.

Konecki, Krzysztof T. (2019). Trust in symbolic interactionist research and in phenomenological investigation. Polish Sociological Review, 207(3), 271-288, https://polish-sociological-review.eu/Trust-in-Symbolic-Interactionist-Research-and-in-Phenomenological-Investigation,117893,0,2.html [Accessed December 11, 2023].

Lakoff, George & Johnson, Mark (1980). Metaphors we live by. Chicago, IL: University of Chicago Press.

LaRoche, Kathryn L.; Jozkowski, Kristen N.; Crawford, Brandon L. & Jackson, Frederica (In press). Can someone be both pro-life and pro-choice? A mixed-methods study of abortion identity labels among US adults. Perspectives in Sexual and Reproductive Health.

Lee, Nam-Jin; McLeod, Douglas M. & Shah, Dhavan V. (2008) Framing policy debates: Issue dualism, journalistic frames, and opinions on controversial policy issues. Communication Research35(5), 695-718.

Logler, Nick; Yoo, Daisy & Friedman, Batya (2018). Metaphor cards: A how-to-guide for making and using a generative metaphorical design toolkit. In Koskinen Ilpo, Youn-kyung Lim, Teresa Cerratto-Pargman, Kenny Chow & William Odom (Eds). Proceedings of the 2018 Designing Interactive Systems Conference. In DIS'18 Designing Interactive Systems Conference, Hong Kong, June 09-13, 2018 (pp.1373-1386). Hong Kong: Association for Computing Machinery.

Mason, Robert; Boersma, Larry & Faulkenberry, G. David (1988). The use of open and closed questions to identify holders of crystallized attitudes: the case of adoption of erosion-control practices among farmers. Rural Sociology, 53(1), 96-109.

Melo-Pfeifer, Sylvia & Chik, Alice (2022). Multimodal linguistic biographies of prospective foreign language teachers I Germany: Reconstructing beliefs about languages and multilingual language learning in initial teacher education. International Journal of Multilingualism, 19(4), 499-522.

Miller, Carrie; Guidry, Jeanine; Dahman, Bassam & Thomson, Maria (2020). A tale of two diverse Qualtrics samples: Information for online survey researchers. Cancer, Epidemiology, Biomarkers & Prevention, 29(4), 731-735.

Nisbet, Eric; Hart, P. Sol; Myers, Teresa & Ellithorpe, Morgan (2013). Attitude change in competitive framing environments? Open-/closed-mindedness, framing effects, and climate change. Journal of Communication, 63(4), 766-785.

Roberts, Brian (2004). Political activism and narrative analysis: The biographical template and the meat pot. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 5(3), Art. 10, https://doi.org/10.17169/fqs-5.3.569 [Accessed: November 27, 2023].

Saad, Lydia (2020). Americans' abortion views steady in past year. Gallup, June, 29, https://news.gallup.com/poll/313094/americans-abortion-views-steady-past-year.aspx [Accessed December 8, 2023].

Satterfield, James & Godfrey, Michael (2011). The University of Nebraska-Lincoln football: A metaphorical, symbolic and ritualistic community event. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 12(1), Art. 3, https://doi.org/10.17169/fqs-12.1.1425 [Accessed: November 27, 2023].

Schmitt, Rudolf (2000). Notes towards the analysis of metaphor. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 1(2), Art. 20, https://doi.org/10.17169/fqs-1.1.1130 [Accessed: November 27, 2023].

Schmitt, Rudolf (2005). Systematic metaphor analysis as a method of qualitative research. The Qualitative Report, 10(2), 358-394, https://doi.org/10.46743/2160-3715/2005.1854 [Accessed December 8, 2023].

Schön, Donald A. (1979). Generative metaphor: A perspective on problem-setting in social policy. Metaphor and Thought, 2, 137-163.

Smithies, Chris & Lather, Patty (1997). Troubling the angels: Women living with HIV. Boulder, CO: Westview Press Inc.

Steele, Rose; Baird, Jennifer & Davies, Betty (2022). Using metaphors to make research findings meaningful. Canadian Journal of Nursing Research54(2), 99-100.

Stets, Jan & Leik, Robert (1993). Attitudes about abortion and varying attitude structures. Social Science Research, 22(3), 265-282.

Tracy, Sarah J. & Malvini Redden, Shawna (2015). Markers, metaphors, and meaning: Drawings as a visual and creative qualitative research methodology in organizations. In Kimberly D. Elsbach & Roderick M. Kramer (Eds.), Handbook of qualitative organizational research: Innovative pathways and methods (pp. 238-248). London: Routledge.

Vadeboncoeur, Jennifer & Torres, Myriam (2003). Constructing and reconstructing teaching roles: A focus on generative metaphors and dichotomies. Discourse: Studies in the Cultural Politics of Education, 24(1), 87-103.

Wheatley, Dawn & Vatoney, Eirik (2022). Understanding attitudes towards social media segregation: Spatial metaphors in the discussion of Twitter blocklists. Information, Communication & Society, 25(1), 1-16.

White, Richard & Green, Anne (2011). Opening up or closing down opportunities?: The role of social networks and attachment to place in informing young peoples' attitudes and access to training and employment. Urban Studies, 48(1), 41-60.


Katie HAUS is an advanced doctoral student working on a dual PhD in health behavior (School of Public Health) and qualitative and quantitative research methodology (School of Education) at Indiana University. She is currently integrating her interests in sexual health with her interests in creative epistemologies and critical methodologies to center human experience and outcomes in reproductive health.


Katie Haus

QQRM, School of Education, Indiana University
201 N. Rose Ave.
Bloomington, IN 47405, USA

E-mail: hauskat@iu.edu

Dr. Jaclyn HADFIELD is a multilingual cross-cultural behavioral scientist who specializes in mixed-methods, cross-cultural, and multilingual research (English/Spanish/Italian). Although in her research she explores various domains of public health, she predominantly focuses on social and behavioral determinants of health beliefs and behaviors among women within the domain of physical activity using theory-based methods with a translational and community-based approach. In her research, she aims to identify salient factors to inclusively influence intervention designs and improve women's physical activity behavioral engagement while reducing health disparities domestically and globally.


Jaclyn Hadfield

School of Kinesiology, Louisiana State University
50 Fieldhouse Dr. Room 214B
Baton Rouge, LA 70803, USA

E-mail: jaclynhadfield@lsu.edu

Dr. Kathryn J. LaROCHE is an assistant professor in the Department of Public Health at Purdue University. As a public health social scientist and applied anthropologist, she carries out community-engaged, action- and intervention-oriented research about sexual and reproductive health with a focus on abortion and pregnancy.


Kathryn LaRoche

School of Public Health, Purdue University
Matthews Hall, Room 227
West Lafayette, IN 47907, USA

E-mail: kjlaroch@purdue.edu

Dr. Barbara DENNIS is a professor in the Qualitative and Quantitative Research Methodology program at Indiana University. She studies methodological concepts such as validity and ethics in the context of critical qualitative research with a particular interest in methodological equity and justice. In her 2020 book, "Walking with strangers. Critical Ethnography and Educational Promise," nominated for several awards, she detailed the methodological inner workings of a long-term critical ethnography. In her most recent book, co-authored with Pengfei HAO, Karen ROSS, and Peiwei LI (2021), she advanced a practitioner/student centered approach to social science methodology.


Barbara Dennis

QQRM, School of Education, Indiana University
201 N. Rose Ave.
Bloomington, IN 47405, USA

E-mail: bkdennis@indiana.edu

Dr. Ronna C. TURNER is a professor in educational statistics and research methods and coordinator for the educational statistics and psychometrics concentration of the interdisciplinary Statistics and Analytics program at the University of Arkansas. She received her PhD in educational psychology from the University of Illinois, specializing in quantitative and evaluative research methodologies.


Ronna Turner

College of Education and Health Professions, University of Arkansas
340 N. Campus Drive
Fayetteville, AR 72701, USA

E-mail: rcturner@uark.edu

Dr. Brandon CRAWFORD is a principal investigator on the Indiana University Abortion Attitudes Project (IUAPP) and an assistant professor in the Department of Applied Health Science in the School of Public Health at Indiana University Bloomington. His main areas of research include the measurement of attitudes regarding controversial social issues and examining the relationship between child maltreatment, child welfare experiences, and adverse life events throughout the life-course. He has expertise in the analysis of secondary data (including data with complex survey designs), survey development, and quantitative analyses.


Brandon Crawford

School of Public Health, Indiana University
501 N. Morton Street, SP 109
Bloomington, IN 47405, USA

E-mail: brancraw@iu.edu

Dr. Wen-Juo LO is an associate professor and program co-coordinator in the Educational Statistics and Research Methodology (ESRM) program at the University of Arkansas. Before coming to the USA, he worked in a general hospital in Taiwan for five years and conducted inpatient and outpatient psychotherapy as well as psychological assessments. His research interests involve methodological issues related to latent factor modeling with a focus on psychometric methods in examining the validity and reliability of the psychological inventory.


Wen-Juo Lo

College of Education and Health Professions, University of Arkansas
340 N. Campus Drive
Fayetteville, AR 72701, USA

E-mail: wlo@uark.edu

Dr. Kristen N. JOZKOWSKI is a principal investigator for the Indiana University Abortion Attitudes Project (IUAPP). She is the William L. Yarber Endowed Professor in Sexual Health in the Department of Applied Health Science in the School of Public Health, a senior scientist with the Kinsey Institute for Research in Sex, Gender, and Reproduction, and affiliate faculty in gender studies at Indiana University. In her research, she focuses on sexual consent and refusal communication and abortion attitudes. She has expertise and training in mixed-methods and sexuality.


Kristen Jozkowski

School of Public Health, Indiana University
237 Sycamore Hall
Bloomington, IN 47405, USA

E-mail: knjozkow@iu.edu


Haus, Katie; Hadfield, Jaclyn; LaRoche, Kathryn J.; Dennis, Barbara; Turner, Ronna C.; Crawford, Brandon L.; Lo, Wen-Juo & Jozkowski, Kristen N. (2024). Doorways of understanding: A generative metaphor analysis [80 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 25(1), Art. 7, https://doi.org/10.17169/fqs-25.1.4097.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Funded by the KOALA project

Creative Common License

Creative Commons Attribution 4.0 International License