Background
The ethical discussion about abortion focuses on two complex issues:
i) if the embryo/fetus is from fertilization or at some stages of pregnancy unequivocally entitled to protection of life and
ii) if the pregnant woman is obliged to allow the embryo/fetus to use her body on some or all occasions [
1]. Because an indisputable ethical conclusion has not been reached, the freedom of conscience in many liberal societies includes regulated abortion. In contrast, anti-abortion advocates regard the fetus as a person from the moment of conception and consider the termination of pregnancy a form of homicide. Principal Christian denominations as well as other major religions oppose induced abortion [
2]. They mostly reserve the right to life-threatening situations of the pregnant woman but are more diverse when it comes to other indications.
The aim of the present study was not to discuss whether it can be morally right or wrong to terminate a pregnancy but to perform a textual analysis on opinions for and against abortion or conscientious objection (CO; refusal by medical professionals to participate in abortion procedures due to religious or ethical reasons). The analysis was conducted for two European countries, Finland and the Republic of Ireland (hereafter, Ireland), that both have quite similar population sizes and health care systems [
3,
4] but very different abortion legislation [
5,
6]. Ireland is considered a morally conservative country [
7] with over 78% of the population identifying as Catholic [
8]. Finland can be considered more liberal [
9], even though 72% of the population belongs to the national Evangelical Lutheran Church [
10]. Obviously, this characterization oversimplifies the situation of the nations and, instead of being uniform, both countries do have voices that are in opposition to the general national ethos.
Abortion remains criminally prohibited in Ireland except in cases where a pregnant woman's life is at risk, including suicidality [
6,
11]. Abortion is not legal in cases of rape, incest or life-limiting conditions of the fetus. A person found guilty of intentionally destroying unborn human life is liable to a possible maximum sanction of 14 years imprisonment, but we were unable to locate any data indicating prosecutions. Every year, thousands of Irish women seek abortion abroad, mainly in the UK, and many others purchase abortifacients from Internet sources [
7,
12]. In contrast, the Finnish health care system grants abortion virtually on demand until the 12th gestational week, but the pregnant woman is obliged to provide justification, why the continuation of pregnancy would be a significant burden (so called “social indications”) [
5,
13,
14]. While at first glance this is different from the more permissive laws of many countries in Western Europe (e.g., Sweden; [
15]), in reality the difference is minor, as practically any justification from the woman is accepted during this period. After the 12th week, the criteria mostly include medical conditions of the woman or the fetus, and social indications are no longer allowed except by the permission of the National Supervisory Authority for Welfare and Health. Between weeks 20–24, only serious medical conditions of the fetus are mentioned as indications. In Finland, 96% of the abortions are performed medically [
16]. Unlike in many other European countries, such as Ireland, there is no CO to participating in abortion procedures in Finland [
6,
13,
17].
To assess the polarized debate on abortion, it would be interesting to examine, how the opposite views are presented in public discourse and how they reflect the thinking patterns of the participants. Generally, there are two interactive information-processing systems: rational and experiential [
18,
19]. The latter is considered evolutionarily old and rapid in everyday situations, where it is crucial to organize and interpret information automatically. Experiential thinking is intuitive and emotional, and its use can cause errors of judgment. This approach prefers concrete information, often in the form of
i) personal or anecdotal experience (testimonials and narratives) [
18‐
20]. Due to
ii) confirmation bias, a person seeks information consistent with previously existing beliefs while alternative hypotheses are not readily considered. In addition, experiential reasoning is characterized by
iii) generalization and stereotypical thinking as well as by
iv) magical beliefs. A morally neutral issue is often given
v) moral significance. Beliefs based on experiential thinking are
vi) resistant to change, and logical evidence and contradictory information have only little influence on them.
Both experiential thinking and fallacies have been suggested to create and enforce false beliefs [
19,
21]. Fallacies are violations of rules for critical discussion, which aims to resolve a difference of opinion [
22]. The evaluation of arguments by an audience can be affected by fallacies, especially in a case of pre-existing biases [
23]. Both experiential thinking and fallacies have a tendency towards generalizations and simplification of complex data [
18‐
20]. In addition, the use of personal experience or narratives, which often take the form of testimonials or appeals to authority, is shared by both, as is the habit of giving moral significance to neutral issues. These similarities provide fertile ground for assessing texts on morally ambiguous issues that are strongly opposed to each other. Regarding the abortion discourse, it is possible that both sides fail to concentrate on available unbiased research data but utilize emotional arguments and experiential thinking. This could cause failure in communication and a stalemate evidenced by the continuous reappearance of the same arguments. It would be of benefit to break this vicious circle of conflict.
The use of experiential and scientific/medical arguments has also been examined previously in abortion debates but with different suppositions [
24,
25]. Data originating from medical sciences have been observed to form a basis for emotional argumentation and persuasion [
26‐
29]. The emotional appeal of ultrasound images, turning of fetal images into a narrative, and appeals to authorities may contain material that originate from science but the actual discourse is not scientific any longer. The present study has a novel approach by making a clear distinction between objective research data and emotional argumentation. This does not mean that emotional arguments would be invalid or that the experiences of people providing the testimonials would not be sincere, but they represent a different type of material, which cannot necessarily be validated from independent sources or by repeating an experiment. Similarly, research data are not always interpreted objectively, but the validity of these arguments can be tested in case of uncertainty, as data based on the scientific method can be returned to or reproduced with the methodology described; and even the validity of the conclusions can be reassessed. Basically, it is always possible to go to the original publication and data to assess if research results were obtained according to the norms of science. In comparison to anecdotal evidence or testimonials, research findings provide a data bank that can be accessed and evaluated. This allows others to do fact checking independent of the people performing the original study—although this is obviously not always conducted in a perfect or objective manner.
Little previous research has been conducted on abortion discourse by politicians vs. laypersons, and the aim of the present study was to identify typical argumentative moves and patterns of experiential thinking in political debate and layperson discussion regarding induced abortion. Two European countries with relatively similar population sizes, governance and health care systems were chosen to see how the cultural and religious differences in Finland and Ireland would influence abortion discussions. If political decision-making were based on non-rational argumentation, it could be difficult to establish legislation applicable not only to one’s peer group but also to those who do not share the same ethical or moral background. As political decision-making should advance the well-being of all citizens, it could be expected of politicians to be able to assess also data that do not support one’s own worldview or preconceptions. It was hypothesized that i) the occurrence of emotional argumentation and features of experiential thinking would be high in abortion discourse as the subject raises strong emotions in people and that ii) parliamentary discussion would contain fewer cases of emotional argumentation and features of experiential thinking than layperson debate.
Methods
The content of 493 online texts and official transcripts of parliamentary debates on abortion was analyzed systematically. The sampled texts represented
i) discussions at the Parliament of Finland (
n = 166, focus on the years 2013–2015, range 2006–2015) and
ii) the lower house of Oireachtas (Dáil Éireann) of Ireland (
n = 122, focus 2013–2015, range 2013–2015) and
iii) online texts in Finnish (
n = 101, focus 2010–2015, range 2005–2016) and
iv) in English (
n = 104, focus 2013–2015, range 2008–2016) from the same countries. These texts were not randomly selected but represented available parliamentary transcripts and online texts obtained by search engines using abortion-related keywords. The transcripts of parliamentary debates were browsed on
https://www.eduskunta.fi/FI/search/Sivut/vaskiresults.aspx and
http://oireachtasdebates.oireachtas.ie/debates%20authoring/debateswebpack.nsf/fulltextsearch?readforms by using the keywords “abortion” and “termination” (“abortti”, “raskaudenkeskeytys” in Finnish) on the websites’ search engines. While the general framework of discussion was the same (abortion), the specific laws being debated differed (Finland: main focus on allowing CO for medical professionals; Ireland: allowing abortion in specific circumstances). Moreover, the Finnish transcripts tended to be shorter than the Irish ones, caused at least partly by more strict time limits. Regarding layperson debates, several types of online texts, such as blogs
, were retrieved by using selected keywords in Finnish and English (“abortion”, “abortion blog”, etc.) with Internet search engines and by browsing platforms representing Finnish and Irish writers. This yielded material from the home pages of private persons and politicians, subscription newspapers, Internet news publications and pro-life/pro-choice organizations. A minority of texts was anonymous. All the selected documents were public, and the authors were not contacted to gain permission for the analysis. The texts were not classified into pro-life and pro-choice, as many authors revealed a position in between these categories.
The analysis of experiential thinking was based on previous literature on the subject [
18‐
20]. The selected features were classified in this study as follows: personal or anecdotal experience (testimonials, narratives and metaphors) as the principal tool to assess data, confirmation bias (seeking information consistent with existing beliefs), stereotyping/generalization (constructing a standardized mental picture of a group that represents an oversimplified opinion/simplifying complex information) and magical/religious beliefs (referring to supernatural phenomena as relevant for the argument, including religious argumentation for secular legislation). Argumentative strategies were spotted in the discursive moves and classified by using several, partly divergent sources for argumentation and fallacies [
22,
23,
30‐
36]. The context and validity of the arguments were assessed and, eventually, an argument was regarded fallacious, if it highlighted aspects that were irrelevant to prove or disprove a claim, for instance, characteristics of a person instead of content that would be pertinent for a writer’s position. The nature of the debate was taken into consideration, as argumentative moves that would be fallacious in some other contexts can be considered sound in political debate [
30,
32]. For instance,
ad consequentiam was regarded as fallacious in cases the alleged consequences of a bill were unsupported by or lacking any evidence, but it was accepted that these examples could be equally well assessed to represent reasonable argumentative strategies in political decision-making [
30,
31]. Detailed descriptions of the fallacies chosen for the analysis are available in the referenced literature [
22,
23,
30‐
36]. Regarding experiential thinking, the feature of attaching moral labels to neutral issues was not included in the analysis as abortion debate was accepted to be a discussion about ethics. In several cases, the classification of arguments and that of experiential thinking overlapped, i.e., the same passage could be assessed to be both fallacious and to contain features of experiential thinking.
Statistical analyses
The prevalence of features of experiential thinking and selected argumentative strategies were calculated by documenting their occurrence in the sampled texts. Multiple occurrences within a text were not recorded due the large variation in text lengths. The distribution of prevalence was analyzed with the χ2 test or, if the test criteria were not met, with the Fisher’s exact test using the IBM SPSS v21.0 program (IBM, Armonk, NY, USA). The p value <0.05 was considered statistically significant. The results are presented as the percentage of texts within a category that contained at least one occurrence of a feature of experiential thinking or an argumentative move.
Statistical comparisons were also performed between the political parties the deputies of which held the most numerous speeches: Christian Democrats (political position: center to center–right,
n = 54), Finns Party (social:
right-wing, economic:
center–left,
n = 39), Social Democratic Party of Finland (center–left,
n = 21), National Coalition Party (center–right,
n = 15) and Left Alliance (left-wing,
n = 14) for Finland and Fine Gael (center–right,
n = 45), Fianna Fáil (center–right,
n = 12), United Left (left-wing,
n = 10) and Labour Party (center–left,
n = 9) for Ireland, which also had one study group of independent politicians (
n = 23). The results were also compared between all center–left-wing and center–right-wing parties, separately for both countries, to see whether political spectrum affected the occurrences of features of experiential thinking or argumentative strategies.