Categorisations of Dis/ability over Time

In this section, we analyse how the Swedish print media names dis/ability and how these naming practices change over time. The results will show which specific linguistic labels have been used, what these labels signal, and how and when new labels replace older ones or how they exist in parallel, which, in turn, may say something about prescribed or acceptable language use at a given point in time.

Table 1 shows the most frequently used comprehensive expressions for dis/ability in Swedish print media. The table includes all inflections of the terms (e.g. of number and definiteness) as well as all compounds with the terms as their first element (e.g. handikapptoalett, 'handicap toilet').

Table 1. Overall terms to name dis/ability in Swedish print media; relative frequency per 10,000 words in brackets.

Term		Frequency
handikapp*	('handicap')	53,111	(20.48)
funktionshind*	('functional hindrance')	19,199	(7.40)
funktionsneds*	('functional deficit')	9,387	(3.62)
funktionsvar*	('functional variation')	640	(0.25)
funktionsförmåg*	('functional ability')	213	(0.08)
funktionsnivå*	('functional level')	72	(0.03)
funktionsstörning*	('functional disorder')	29	(0.01)
funktionsgrad*	('functional grade')	17	(0.007)
normbrytande funktionalitet*	('norm-breaking functionality')	4	(0.002)
funktions-förutsättning*	('functional condition')	3	(0.001)
funktionstillstånd*	('functional condition')	2	(0.0008)
funktionsskillnad*	('functional difference')	1	(0.0004)

By far the most frequently used term to name dis/ability in Swedish print media is handikapp* ('handicap'). With its 53,111 occurrences, it is used nearly twice as often as all the other terms taken together. Funktionshind* ('functional hindrance') is the second most used term with 19,199 occurrences and funktionsneds* ('functional deficit') with 9,387 occurrences is the third most used. The other terms all occur fewer than 650 times. Some of them, like normbrytande funktionalitet* ('norm-breaking functionality') and funktionstillstånd* ('functional condition'), occur fewer than a handful of times.

All in all, the overview of the existing terms in Table 1 also shows that all terms, besides handikapp ('handicap'), include the word funktion ('function'). Thus, an explicit reference to someone's functionality appears to be an important aspect in mentions of dis/ability in Swedish.

As mentioned above, the term funktionshinder ('functional hindrance') is meant to denote a limitation that an impairment imposes upon a person in relation to the environment rather than a lack of ability on an individual level, at least according to the official definitions of the National Board of Health and Welfare from 2007. In order to gain a fuller understanding of the use of funktionshinder in our corpus, we studied the meaning of the term in 300 occurrences, which were randomly selected by the corpus analysis tool. The analysis shows that the term is not used in any of these occurrences to refer to the relationship between individual/group and environment, whether in the texts published before 2007 or after. Rather, throughout the whole sample it specifically refers to a lack of ability on an individual level, often expressed through phrases such as person med funktionshinder ('person with functional hindrances') or att ha ett funktionshinder ('to have a functional hindrance'). In other words, the relationship between the individual and the environment, which is part of the definition of funktionshinder, is not reflected in the use of the term.

The term funktionsvariation ('functional variation') was introduced by disablised activists with the aim of challenging the distinction between disability and ability by denoting the varying levels of functional ability that can be found within the whole population. However, our analysis of 300 random examples from the corpus shows that it is used to exclusively name disablised positions in 293 of the examples. The remaining seven examples consist of metalinguistic explanations of the term's norm-critical denotation. Five of these seven occurrences appear in direct quotes of disablised activists. Thus, funktionsvariation ('functional variation') is on the whole not used as intended by activists but synonymously to the more established forms as a denomination of a non-normative position, at least in print media.

As Figure 1 illustrates, the usage of the four most frequent terms is unevenly spread over time. The y-axis shows the relative frequency of usage of the terms per 10,000 words, which makes the different sized corpora from the different years comparable.

Figure 1 shows that the usage of the different terms is separated into distinct phases. From the 1980s until 2016, handikapp* ('handicap') is the most common term. Funktionshind* ('functional hindrance') appears at the beginning of the 1990s, and as of 2006, funktionsneds* ('functional deficit') is being increasingly used. The emergence of funktionsneds* ('functional deficit') in the press corresponds to the language policy action by the National Board of Health and Welfare at the end of 2007. Since 2015, funktionsvar* ('functional variation') occurs as a fourth alternative in print media but is clearly used the least frequently.

In 2019, funktionsneds* ('functional deficit') was the most common term used in connection to disability within Swedish newspapers, but the other three terms were also in use, with, notably, handikapp* ('handicap') decreasing and funktionsvar* ('functional variation') increasing.

One can see the frequency of usage of handikapp* ('handicap') decreases almost constantly during the studied period, implying that the term is gradually being replaced by the other terms. However, added together the frequency of usage of all four terms at the end of the period (2019) is around thirty usages per 10,000 words, which does not correspond to the frequency of usage of handikapp* ('handicap') in the 1980s and 1990s (between thirty and ninety usages per 10,000 words). This may partly be explained by an increasing length of the articles. However, this question is not further explored here.

All in all, this first analysis shows that there are several different terms for dis/ability in use in the Swedish press. All of these terms are nearly without exception used to name non-normative positions by focusing on a lack of ability on an individual level. The perspective of dis/ability as a relationship between the individual and the environment is absent. Furthermore, the analysis shows a shift from the sole use of handikapp* ('handicap') to a usage of compound words with funktion ('functional'). The usage of terms in print media thereby adheres to language change "from above", especially to the recommendations from the National Board of Health and Welfare, when it comes to replacing handikapp* ('handicap') with other terms, albeit with a delay of some years and without adapting the intended meanings of the terms.

Importantly, these results indicate how difficult it may be to change how media, and by extension, mainstream society, talk about and understand dis/ability. Firstly, handikapp* ('handicap') is still in use although it has been strongly discouraged for a long time. Secondly, the two terms which have been introduced in order to introduce new ways of understanding dis/ability: funktionshind* ('functional hindrance', which attempts to identify the role of the environment) and funktionsvar* ('functional variation' which challenges the distinction between disability and ability and enables the identification of variation at group and societal levels) have both instead ended up being used to indicate a property of an individual, along a medical-biological model (e.g. 'person with a functional variation').

Categorisations of People

We now turn to linguistic practices of categorising people through disablising and ablising naming practices, it should be reiterated that by labeling someone, we simultaneously categorise that person. Similarly, the very act of naming groups of people in a sense creates these groups and the distinctions between the groups. Thus, it is important to understand how both normative and non-normative groups are created through naming practices. One means of investigating how people are disablised through language is through an examination of people-first language in contrast with identity-first language. Examples of these include barn med funktionshinder ('children with functional hindrances') which is used in contrast with funktionshindrade barn ('functionally hindered children').

The search terms that we have selected are based on the terms that were found to be the most commonly used in the previous section. This yielded the following search terms for people-first language: med handikapp ('with handicap/s'), med funktionshinder ('with functional hindrance/s'), med funktionsnedsättning/ar ('with functional deficit/s'), and med funktionsvariation/er ('with functional variation/s'). In comparison, for identity-first language, we selected the corresponding search terms handikappad ('handicapped'), funktionshindrad ('functionally hindered'), funktionsnedsatt ('impaired'), and funktionsvarierad ('functionally varied'), with all their inflections. 2

After manually filtering non-relevant search hits, a total of 6,597 instances of people-first usage and 30,084 of identity-first usage were found in our corpus of 56,666 articles. Thus, throughout the material as a whole, identity-first usage is much more common than people-first usage. However, it should be noted that the identity-first examples include not only adjectival uses, as in funktionshindrade barn ('functionally hindered children'), but also nominal uses so that e.g. funktionshindrad on its own corresponds to a noun phrase such as 'a functionally hindered person'.

Table 2. People-first and identity-first usage in Swedish print media.

People-first usage		Identity-first usage (adjectival and nominal usage)
med handikapp ('with handicap/s')	494	handikappad ('handicapped')	18,113
med funktionshinder ('with functional hindrance/s')	2,484	funktionshindrad ('functionally hindered')	10,373
med funktionsnedsätt-ning/ar ('with functional deficit/s')	3,332	funktionsnedsatt ('impaired')	1,552
med funktionsvariation/er ('with functional variation/s')	287	funktionsvarierad ('functionally varied')	46
Total	6,597	Total	30,084

Bearing this in mind, while looking more closely at the different individual terms, Table 2 shows that identity-first adjectival and nominal usage of handikappad ('handicapped') occurs much more frequently (18,113) than people-first usage med handikapp ('with handicap/s') (494). The same pattern is seen for 'functional hindrance', with identity-first adjectival and nominal use of funktionshindrad ('functionally hindered') being more frequent (10,373) than people-first usage, med funktionshinder ('with functional hindrance/s') (2,484). The opposite pattern is instead found for the two remaining terms. For 'functional deficit', people-first use of med funktionsnedsättning/ar ('with functional deficit/s') is more common (3,332) than identity-first usage, funktionsnedsatt ('impaired') (1,552). For 'functional variation' people-first use of med funktionsvariation/er ('with functional variation/s') (287) is much more common than identity-first usage, funktionsvarierad ('functionally varied') (46).

As we saw in Figure 1, there is a clear chronological order of appearance in Swedish print media for these terms. Thus, over time, we find a pattern of people-first usage gaining more ground, first being in clear minority use in the case of 'handicap' and 'functional hindrance', and then becoming the main form of usage for 'functional deficit', which began to be increasingly used from 2008 onwards, and even more so for the most recently introduced term, 'functional variation', which began to appear in 2014–2015.

Another way of understanding naming practices is achieved by looking at who is being disablised, or, in other words, which other categorisations are being made conjointly with disablising naming practices. We here investigate this in relation to people-first usage. We have manually investigated all one-word positions occurring immediately before med handikapp, funktionshinder, funktionsnedsättning/ar, funktionsvariation/er ('with handicap/s, functional hindrance/s, functional deficit/s, functional variation/s'), and classified them into thematic categories based on the data.

Figure 2 indicates that over half of all people-first instances in the data are used with general, unspecified categorisations such as personer ('people'), människor ('people', 'human beings'), många ('many'). This is the case for fifty-three percent of all examples of med handikapp ('with handicap/s'), increasing to 59 percent for med funktionshinder ('with functional hindrance/s'). In the case of med funktionsnedsättning/ar ('with functional deficit/s') and med funktionsvariation/er ('with functional variation/s'), it corresponds to around three quarters of all examples–72 and 73 percent respectively. Thus, an increase in people-first usage in print media seems to be accompanied by a proportionally increased media coverage of disablised people as a generalised group of people.

The second-largest thematic category to occur with people-first usage is age, with children being the most frequently mentioned age group. Examples include barn med handikapp ('children with handicap') and barn och ungdomar med funktionsnedsättning ('children and young people with functional deficit'). Categorisations of age account for 19 percent of all people-first examples in the material as a whole, with 30 percent of all examples of med handikapp ('with handicap/s') and 13 percent of all examples of med funktionsvariation/er ('with functional variation/s').

Other thematic categorisations are comparatively minor. They are the following, ordered according to frequency:

• work/occupation: anställda med funktionshinder ('employees with functional hindrances')
• sports: idrottare med funktionsnedsättning ('athletes with functional deficit')
• we and they: oss med funktionsvariation ('us with functional variation')
• gender: kvinnor med handikapp ('women with handicap')

It should specifically be noted in relation to Figure 2 that "we and they" categorisations are proportionally larger for med funktionsvariation/er ('with functional variation/s') than for any of the other terms. This may reflect the sometimes controversial use of this relatively newly coined term (Sandström 2016) and that the term is used in argumentations about the term itself.

Investigating naming practices regarding normative positions is less straightforward than investigating non-normative positions, as normative positions typically are not named at all (Hornscheidt 2012; Kitzinger 2005; Wojahn 2015). As Goodley (2009, p. x) puts it: "normality and normalcy is achieved through an unsaying: an absence of descriptions of what it is to be normal." For instance, a statement about "children" may carry with it a tacit understanding of these children as belonging to a normative position regarding dis/ability unless otherwise specified, without this ablising being overtly expressed through language. It can be assumed that the vast majority of portrayals of ablised positions in the material are of this covert kind.

However, as a first step, we investigate the frequency of the antonyms of the terms that are used for disablising naming practices. That is, with regards to people-first usage–med handikapp ('with handicap/s') etc.–we have used the corresponding utan handikapp ('without handicap/s') and so on, as search terms. For identity-first usage– handikappad ('handicapped') etc.–we have used the corresponding negated terms, icke-, icke, inte, ej handikappad ('non-/not handicapped').

In terms of numbers, explicitly referenced normative positions using either utan ('without') or negation are considerably less frequent than the corresponding disablising naming practices. As seen above, there are in total 6,597 examples of people-first uses in the data. The corresponding utan handikapp, funktionshinder, funktionsnedsättning/ar, funktionsvariation/er ('without handicap/s, functional hindrance/s, functional deficit/s, functional variation/s') occur only 262 times. Regarding identity-first usage, the 30,084 instances of disablising naming practices that we saw above, are matched by a total of 231 examples of icke-, icke, inte, ej handikappad/funktionshindrad/funktionsnedsatt/funktionsvarierad ('not handicapped/functionally hindered/impaired/functionally varied'). Typical examples are the following:

Fem handikappade barn och minst lika många barn utan handikapp ska vara med, när verksamheten byggts ut. (Tidningarnas Telegrambyrå, 18 November 1991)

'Five handicapped children and at least as many children without handicap are to participate, when the centre is to expand.'

- I dag är Internet viktigare för de funktionshindrade än för de icke-funktionshindrade, säger NN. (Tidningarnas Telegrambyrå, 22 January 2001)

'- Today Internet is more important to the functionally hindered than to the non-functionally hindered, says NN.'

In these two examples, disablising and ablising naming practices both occur, with a clear distinction between the two naming practices and between the two groups of people which are being linguistically created by these practices. Thus, such term pairs are clearly used to demarcate two separate groups of people.

The numerical differences show that explicitly ablising naming practices are much less frequent than disablising naming practices, at least when it comes to antonyms expressed with utan ('without') or negations. But the press texts contain further linguistic ways through which ablised positions are created. In many texts, the authors use specific adjectives to name ablised positions, for example, frisk ('healthy'), vanlig ('ordinary'), and normal ('normal'), as in friska barn ('healthy children'), vanliga människor ('ordinary people'), and normala elever ('normal pupils'). The following two quotes exemplify the use of frisk ('healthy') and vanlig ('ordinary'):

[Ä]r badet numera mest till för barnfamiljer utan handikapp? Vi som behöver varmbadet och får hjälp ner och upp och har personliga assistenter, vi får inte plats och blir dränkta av lekande friska barn. (Mitt i Söderort Hammarby-Skarpnäck, 29 July 2014, p. 2)

'[I]s the bath nowadays mostly for families with children without handicap? We who need the hot bath and receive help to get in and out and have personal assistants, there's no room for us and we're being drowned by playing healthy children.'

[…] vanliga resenärer likväl som handikappade skulle tillfredsställas. (Svenska Dagbladet, 6 July 2001, p. 48)

'[…] ordinary travellers as well as handicapped would be satisfied.'

In the first example, friska barn ('healthy children') is an ablising naming practice, with friska ('healthy') being used as a synonym for utan handikapp ('without handicap') earlier in the extract. The second example illustrates how naming practices are used to categorise people into two mutually exclusive groups. Here, the disablising handikappade ('handicapped') is contrasted with the ablising vanliga resenärer ('ordinary travellers'). There are also some uses that blur or challenge binary distinctions between ablising and disablising naming practices. For instance, consider the following example:

Fortfarande är det nog så att många ser oss med funktionshinder som ett släkte för sig, inte som vanliga människor med funktionshinder. (Göteborgs-Posten, 6 July 1995, p. 4)

'Even today there are probably many who see us with functional hindrance/s as a race apart, not as ordinary people with functional hindrance/s.'

The juxtaposition of vanliga ('ordinary') with med funktionshinder ('with functional hindrance/s') in this argumentative sentence constitutes a naming practice which resists being either simply ablising or disablising, in contrast with the other examples we have seen so far. Other examples of naming practices that blur binary distinctions are compound words which combine a word that usually is used in an ablising way with a word that is used in a disablising and often derogative way, for instance, normalstörd ('normally disordered'). Expressions that blur or challenge binary distinctions between ablising and disablising naming practices occur mostly in quotes by disablised activists or in self-descriptions of organisations working with disablised people.

Summing up this section and starting with disablising naming practices, it can be seen that people-first usage increases over time, and for the most recently introduced terms, people-first usage is more common than identity-first usage. People-first usage is predominantly used with general references, and this increases with the most recently introduced terms. After 'general references', 'children' is the second-largest categorisation to be included in people-first expressions, and other categorisations are comparatively minor. Ablising naming practices are more difficult to identify using corpus-linguistic methods and can be assumed to be done mostly tacitly, without dis/ability being named at all. Ablising naming practices using negation or utan ('without') are considerably less frequent than disablising naming practices.

Importantly, these results show that Swedish print media clearly disablises people through specific linguistic labels, thereby creating a dichotomy between people who are being disablised and those who are not. Furthermore, the results suggest that people who are ablised are rarely done so through explicit reference to dis/ability. In other words, when people are ablised tacitly, without labels such as utan handikapp ('without handicap'), the ablising that is being done can remain unnoticed and normalised. By stating this, we do not intend to argue in favour of an increased use of labels such as utan handikapp ('without handicap'), but rather to point out a need for making ablising processes explicit.

The Thematic Contexts of Categorisations

To this point, the focus has been on an analysis of which overall terms are used to name dis/ability and how categorisations of people as 'ablised' and 'disablised' are linguistically realised. In this section, we proceed to an investigation of what the Swedish press is writing about dis/ability. In order to accomplish this, there is a need to analyse the words that regularly appear near the terms that name dis/ability. These frequently co-occurring words are called "collocates" and contribute to the meaning of the word they occur next to or near to (the node). The analysis of a node's collocates can then provide a "helpful sketch of the meaning/function of the node within the particular discourse" (Baker et al. 2008, 278).

Thus, a collocation analysis will enable us to study which topics the Swedish media discusses in relation to disability. Since media texts are helping people to build up a normative view of the world it is important to study how dis/ability is framed in these widely read texts. In the previous sections, it was demonstrated that the terms referring to disability have changed over time. However, there is a corresponding need to investigate whether changes in the naming practices go along with changes in how dis/ability is framed; that is, whether the new terms also mean a new understanding of dis/ability.

The collocation analysis starts with an analysis of the collocates of funktionsnedsättning ('functional deficit'), which is the most frequently used term in Swedish print media to name dis/ability today, as we saw earlier. Consequently, the results of the collocation analysis of funktionsnedsättning ('functional deficit') need to be, in a summary fashion, compared to those of handikapp ('handicap'), funktionshinder ('functional hindrance') and funktionsvariation ('functional variation').

The word funktionsnedsättning ('functional deficit') occurs 5,074 times in the corpus. Figure 3 shows the collocates of this word, with funktionsnedsättning ('functional deficit') as the node in the middle, and the collocates around this node. The words to the left of the node are most frequently used before the word funktionsnedsättning ('functional deficit'), and the words to the right are most frequently used after it. The closer a collocate is to the node, the stronger the association between the term funktionsnedsättning ('functional deficit') and the collocate. And the darker a collocate's shade of colour, the more frequent the collocation. As an example, the word personer ('persons') appears with a dark colour to the left of the node, meaning that personer ('persons') most often occurs before funktionsnedsättning ('functional deficit') and that it is a highly frequent collocation. The collocate personer ('persons') appears at a middle distance from the node, indicating that the association between the two words is fairly strong, but there are other word pairs whose association is stronger.

The graph in Figure 3 shows all 152 collocates, which occur more than fifteen times within five words before or after the term funktionsnedsättning ('functional deficit').

The collocates in Figure 3 can be grouped into several categories. In consultation with all authors, we conducted an inductive categorization based on established corpus linguistic procedures (Baker 2023, 135–160) and identified six thematic contexts:

• denominations of persons
• specifications of impairment
• references to discrimination
• care and governmental responsibility
• labour market
• sport

The most common type of collocates are words that name people and, in some cases, categorise them regarding age and gender. Typical collocates of this group are personer ('persons'), människor ('people') and barn ('child/ren'). They are usually used in phrases like personer med funktionsnedsättning ('persons with functional deficit'). We analysed such people-first naming practices in the previous section.

The second-largest group of collocates consists of words which specify types of impairment. Most of these collocates are adjectives, which usually are used directly before the term funktionsnedsättning ('functional deficit'). The most common combination in the corpus is psykisk funktionsnedsättning ('psychological functional deficit'). Other frequently used specifications are intellektuell ('intellectual'), neuropsykiatrisk ('neuropsychiatric') and fysisk ('physical').

Swedish print media also uses the term funktionsnedsättning ('functional deficit') frequently in relation to discrimination, which forms a third group of collocates. Firstly, there is a group of collocates which name different acts of discrimination against disablised people, e.g. våld ('violence'), diskriminering ('discrimination') and hinder ('hindrance'). Secondly, there is a group of collocates that name other types of structural discriminations, most frequently religion ('religion'), ålder ('age') and kön ('sex/gender'). Like in the following quote, they are often used in longer enumerations:

[H]ar du känt dig särbehandlad på grund av könstillhörighet, ålder, trosuppfattning, funktionsnedsättning? (Bohusläningen, 29 July 2013, p. 8)

'Have you felt discriminated against because of gender, age, beliefs, functional deficit?'

The discriminating structures which are named in such enumerations are generally included in the Swedish anti-discrimination law, and funktionsnedsättning ('functional deficit') is usually named as the last or second to last one.

A fourth group of collocates contains words that relate to care and governmental responsibility. The collocates referring to care, concern, for instance, diseases (ohälsa 'illness'; sjukdom 'disease'), assistance (assistans), care (omsorg 'care'; sjukhem 'nursing home'; patient 'patient') and elderly care (äldreomsorg). They are often framed within contexts of political and economic decisions around these items (Socialstyrelsen 'National Board of Health and Welfare'; LSS 'Support and Service for Persons with Certain Functional Deficits', kostade 'cost').

The fifth group of frequent collocates of funktionsnedsättning ('functional deficit') consists of words that refer to the labour market. Through these words, what is mainly expressed is a political need or request that disablised people must get employment (jobb 'job'; sysselsättning 'employment'; arbetsmarknaden 'labour market').

The last, slightly smaller group contains words that name people in sports (idrottare 'athletes'; spelare 'players'; längdskidåkarna 'the cross-country skiers').

Comparing these results to analyses of the frequent collocates of handikapp ('handicap'), funktionshinder ('functional hindrance') and funktionsvariation ('functional variation'), respectively (not shown as figures in this article), we see that there are nearly no differences. The commonly co-occurring words for these three terms are almost exactly the same as for funktionsnedsättning ('functional deficit'), and the same thematic groups can be identified. Handikapp ('handicap') shows somewhat more collocates referring to sport and funktionsvariation ('functional variation') shows comparatively more references to the labour market (e.g. anställa 'employ'; arbetsliv 'working live'). Apart from these minor variations, there are no major differences.

Importantly, the results show that the different linguistic labels which are used to refer to dis/ability in Swedish print media during the last four decades are part of very similar patterns when it comes to co-occurring words. As commonly co-occurring words tell us something about the meaning of a word, it can be assumed that the different Swedish linguistic labels that are used to refer to dis/ability, more or less mean the same thing. Additionally, as these linguistic labels have been introduced at various points in time, these similarities of meaning may indicate that the change of labels over time has not involved a change of meaning. In other words, the findings in this section further strengthen the earlier indications in our results, that a change of linguistic labels does not necessarily translate into accompanying changes to understandings. In fact, the opposite seems to be true in this study–new labels end up taking the same meaning as the old labels, thwarting hopes for new ways of talking about and understanding dis/ability.