Even though there are strong indications of an elevated incidence of language problems in prison populations (Bryan, 2004; Bryan et al., 2007; Snow and Powell, 2011; Morken et al., 2021), awareness of language needs is not high in the justice system, and assessment and help are generally not readily available. There is a need for more knowledge about the language problems of people in prison to better understand how language services may benefit this group both prior to, during, and after incarceration. This paper explores associations between language and mental health in female prisoners.

Prisoners are a vulnerable population, and female prisoners particularly so. Women in prison are more likely than men to experience chronic medical disorders, psychiatric disorders, and drug dependence (Binswanger et al., 2010). Many come from very disadvantaged backgrounds of abuse and addiction, while their crimes are most often associated with their life situation and few can be considered a danger to society (Fair, 2009, p. 3). World-wide, the proportion of women and girls in prison is low. Recent numbers estimate the population at 6.9% of the total number of prisoners (3.3% in Africa, 5.9% in Europe, 6.7% in Oceania, 7.2% in Asia, and 8.0% in the Americas) (Fair and Walmsley, 2022, p. 2). Because of the low numbers of women, prisons tend to prioritize the needs and requirements of male prisoners (Fair, 2009). Challenges facing women in prison include older buildings that often cannot offer adequate conditions for physical activity and fresh air, work and spare-time activities that have low priority because of security issues and lack of resources, and health services that are often less available for women than for men. Furthermore, there can be problems related to safety, especially for women serving in mixed prisons (Falkanger, 2016; Jones, 2019). However, even though the proportion of women in prison compared to men is low, there are at any given time more than 740.000 individual women and girls in prison globally (Fair and Walmsley, 2022), underscoring the necessity and importance of addressing the needs of this group. Adapting prison services to the needs of female prisoners is required to assist in their rehabilitation and reentry into society (Bartels and Gafney, 2011; Falkanger, 2016). Still, with a few exceptions, studies of the prison population tend to include mainly male participants. This is perhaps natural, given the proportion of male to female prisoners, but knowing that the characteristics and needs of women in prisons often do not align with those of men it should be evident that more knowledge about these women is necessary. One thing we do know is that even in the general population rates of depression and anxiety are significantly higher among women than men (Cyranowski et al., 2000; Pigott, 2003; Faravelli et al., 2013; Asher et al., 2017), and that in prison an even larger proportion of women suffer from mental health problems (Svendsen et al., 2023).

Mental health can be defined in at least three ways: as absence of disease, as a state that allows the organism the full performance of all its functions, and as a state of balance within the person and between the person and the physical and social environment (Sartorius, 2002). Poor mental health can, in turn, have adverse effects on quality of life (Spitzer et al., 1995; Meule and Voderholzer, 2020) and work-force participation (Frijters et al., 2010), which will also impact possibilities for making an income. Previous research has shown that people with language problems display more symptoms of depression, anxiety, behavioral problems, and difficulties with attention than those without language problems (Beitchman et al., 2001; Conti-Ramsden et al., 2013; Botting et al., 2016; Helland et al., 2022b). Helland et al. (2022b) found that girls with language problems were more likely to experience emotional problems, whereas boys were more likely to display behavioral problems. Moreover, language problems can often be misunderstood by the environment as boredom, evasion, or resistance (Snow and Powell, 2004a; Helland et al., 2014b), and can lead to avoidance of social interaction, aggressiveness, negative behavior, and low self-esteem (Hopkins et al., 2016). For the prison population, Hughes et al. (2017) found that having language needs was associated with a higher risk of self-harm and substance abuse.

In this paper, we use the term (oral) language problems to refer to any self-reported difficulties with oral language This means that we do not exclusively refer to persons who qualify for a diagnosis of developmental language disorder (DLD) (Bishop et al., 2017), but rather include a broader range of experienced difficulties in oral language production and comprehension. This is both because opportunities for diagnostic testing were not available at the time and because rates of diagnosis have historically been low. Hence, even though DLD has a population prevalence of around 10% (Norbury et al., 2016), when asking participants through self-report if they have been diagnosed with DLD, it is likely that they have not, even in cases where they do indeed fulfill criteria for a diagnosis. Furthermore, relating to mental health the beliefs a person has about their language ability, and their self-efficacy related to language, may be just as important as results of objective measures. If you feel you have a language problem, that may have a stronger impact than in fact having a language problem and not being aware of it. Self-report may mean both overestimation and underestimation of actual problems (Young et al., 2015), but in our context the most important point is experienced problems. Because of the considerations mentioned, we wanted to stay rather wide in scope, and first and foremost focus on experienced problems with oral language.

Having speech and language problems can be difficult not only in terms of general life skills and interpersonal relationships. It can also pose significant challenges for legal protection. Navigating the justice system requires good language comprehension and good expressive skills, for example in forensic interviews and in court proceedings (Snow et al., 2012; Hopkins et al., 2016; Hughes et al., 2017). In an interview study with 31 mainly male young offenders, the majority pointed out that communication abilities were important when presenting in court, and several believed that it could also affect severity of punishment (Hopkins et al., 2016). The bulk of research into these issues, has been done on young offenders, but there is no reason to expect the situation to be significantly different for adults.

Language ability is important from the very first contact with law enforcing authorities where there is often a need to both understand complex language and communicate appropriately (Rost and McGregor, 2012; Snow, 2019). Moreover, there is evidence that listening comprehension is directly related to comprehension of legal rights (Lieser et al., 2019). For example, Lieser et al. (2019) showed that 75% of a group with language disorder did not understand when being read their Miranda rights (the “reading of rights” that is done by police officers in the United States upon arresting someone) compared to 30% in a control group. Furthermore, they did not oversee the consequences of waiving these rights. Naturally, it is difficult to act appropriately if you do not fully comprehend the situation you are in. In the context of court proceedings, Johnston et al. (2016) found that in their sample, many had problems with both core terminology and with understanding court procedures and the roles of the different participants in the proceedings, such as the jury, the witness, the prosecution lawyer, and even the defendant. Notably, there is also evidence that having oral language problems can severely impact prognosis, with young offenders with DLD being more than twice as likely to reoffend as those without language issues (Winstanley et al., 2021). Thus, it is clear that language problems can have great impact on several life arenas, from dealing with the justice system to academic achievement and mental health. Importantly, Hopkins et al. (2016) found that their participants, who were young offenders, were themselves aware of their challenges in communication and literacy. More than half expressed concerns about their language and literacy abilities, and identified clarity of speech, aggression, swearing, and confidence in conversing with others as potential candidates for improvement in the oral domain. This partially reflects that language is a multi-faceted phenomenon which includes several domains and subskills, ranging from the processing and production of the speech-sound signal to complex social communication, and it is possible to have challenges in one or more of these domains. Still, most studies compare impaired and non-impaired groups, without addressing the dimensionality of that impairment. Hence, it is not clear how mental health relates to these different language domains.

However, in a sample of children referred to an outpatient child and adolescent psychiatric clinic, Brenne and Rimehaug (2019) found pragmatic language skills especially to be strongly associated with mental health factors. In their clinical sample, pragmatic skills were lower, and rates of pragmatic language impairment (PLI) higher compared to community samples. Significant correlations were found between pragmatic skills and anxiety/depression, withdrawal, social problems, thought problems, and attention problems. However, they did not find correlations between mental health measures and speech, syntax, social relationships, or special interests. The authors concluded that PLI seems highly comorbid with mental health problems.

As mentioned, most of the research into oral language in prison inmates, has been done on young offenders, with few studies on adults (Morken et al., 2021). Studies generally report a significantly elevated prevalence of speech and language problems compared to the general population. Snow and Powell (2011) used a rather strict criterion to identify language impairment, and 37% of their sample still fell into this category. Bryan (2004) studied different language components in a group of young offenders and found that the number of participants scoring significantly below what is expected for their age differed between language domains (naming 43%, grammatical competency 73%, language comprehension 23%). Other studies found elevated rates of difficulties in prison populations within other language domains, for example figurative and abstract language (Snow and Powell, 2004b), receptive grammar (Bryan et al., 2007), receptive language in general (Hughes et al., 2017), and narrative abilities (Snow and Powell, 2004b). Research has also shown that language impairment and attention deficit hyperactivity disorder (ADHD) often co-occur (Cohen et al., 1998; Tirosh and Cohen, 1998; Sciberras et al., 2014; Helland et al., 2014a, 2016), and that there is increased prevalence of both language problems and ADHD in the prison population (Helland et al., 2022a). It has been suggested that difficulties with pragmatic language skills can explain elevated rates of social difficulties in persons with ADHD (Staikova et al., 2013).

Altogether, this underscores the necessity of exploring the impact speech and language problems can have on prisoners so that prisons can offer the best possible services to assist in rehabilitation. To our knowledge, no previous studies have examined the associations between different language domains and mental health in this population.

The association between language and mental health is well established (Conti-Ramsden et al., 2013; Botting et al., 2016; Hughes et al., 2017; Helland et al., 2022b), and there are several reports of an elevated incidence of both language problems (Bryan, 2004; Bryan et al., 2007; Snow and Powell, 2011) and mental health problems (Hatton et al., 2006; James and Glaze, 2006) among prisoners. This study sought to explore this association in women in prison, since most previous studies have largely focused on men. Our main findings were that there was a clear connection between overall language and overall mental health. When going into detail, we found pragmatics to be the main driver of this effect. Language production and comprehension showed associations with anxiety only. We also found that though there was no difference in mental health between those scoring above and below cut-off for general language problems, the group with possible pragmatic impairment showed poorer mental health than those without. Conversely, there was no difference in general language skills between the groups scoring within and outside the range of psychological distress, but those with psychological distress evaluated their pragmatic skills as significantly poorer than those scoring outside the clinical range for mental health. This indicates that pragmatics can be central to the association between language and mental health in women in prison.

This is in line with previous findings by Brenne and Rimehaug (2019) who also found pragmatics to have stronger associations with mental health than other language domains, albeit in a different population. Likewise, Cohen et al. (2013) found that as much as 45% of a group of youth referred for assessment and treatment at a mental health center had impairment in higher order language function, compared to 15% of controls. Previously, Snow and Powell (2004a) have pointed to poor social and abstract language skills in young offenders. It is, however, difficult to say which direction the association between these higher order language skills and mental health takes. Most probably the influence is bidirectional. One can speculate that as far as the person is aware of their poor pragmatics skills, this can potentially contribute to insecurity in social situations, which in turn can impact self-esteem and mental health. Notably, since this study is based on self-report, the participants’ self-evaluations are the basis of analysis, meaning that only persons who are indeed aware of their difficulties would show up in the group with suspected impairment. Assessment by close others could complete this picture, but due to ethical regulations this was not possible in this project. On the other hand, mental health problems are known to be associated with being more self-critical (Iancu et al., 2015), meaning that persons with mental health problems possibly evaluate their skills more strictly than those in the typical range.

This resonates with the finding that the subscale of anxiety stood out by showing associations with language production and comprehension in addition to the association with pragmatics. We know that anxiety is negatively correlated with self-esteem and self-efficacy, and positively correlated with dependency and self-criticism (Iancu et al., 2015). Hence, it is conceivable that this effect is rather a manifestation of this group being more unforgiving of themselves and less likely to evaluate their communication abilities positively. In other words, this could be a result of differences in subjective assessment, rather than reflecting objectively poorer skills. This could be clarified by including structured individual language assessment in future studies.

However, direct testing of pragmatic skills is not necessarily straightforward, and the ecological validity of such testing has been questioned (O’Neill, 2007). Pragmatics, or social communication, involves many different subskills, and is extremely context dependent. Direct testing of pragmatic skills often involves either pictures or figurines that are used to depict different scenarios, where the person being tested is asked to somehow interpret and respond to the situation that is shown. This is at best a proxy for real-life social situations, and it is not given that the responses a person gives in a clinical setting would match their behavior in real life. Hence, pragmatic impairments are often more evident in real-life situations than in clinical testing (Botting, 2004). Therefore, it is not uncommon to use questionnaires to close others and/or to the person themselves to assess these skills (O’Neill, 2007; Helland and Møllerhaug, 2020).

When it comes to interventions for adults with pragmatic difficulties, the body of literature is small, and mostly related to groups with either acquired brain injuries or with progressive neurological diseases. In a review of interventions for adults with traumatic brain injury, Finch et al. (2016) found that interventions for social communication skills were generally beneficial. Moreover, they found indications that especially context-sensitive approaches were effective, and more so than the more traditional impairment-specific approach which targets more or less exclusively the damaged function. Context-sensitive approaches, on the other hand, combine impairment-based interventions, functional activities, and context-supported participation in a holistic treatment aimed to improve everyday function (Ylvisaker, 2003). There is great need for more research into interventions for adults with pragmatic impairments without known etiology.

An important side note is that Norwegian prison services are based on the principle of normality, which in short implies that the punishment lies in the restriction of liberty itself. Therefore, people who are imprisoned have the same rights as the rest of the population, and their living conditions in prison should as far as possible resemble those in the rest of society. Furthermore, the whole justice system is explicitly built on humanitarianism, due process of law, and equality of treatment (Department of Justice and Police, 2008). This means that though having your freedom taken away may of course in itself have consequences for your mental health, the effects we see should not be caused by adverse treatment or living conditions. Still, we know that appropriate treatment for mental health problems may not be readily available in prisons, and perhaps even less so than in general society. Moreover, we do not know if the reported psychological distress was present before incarceration, or if it has appeared during the prison stay.

In sum, there seems to be a strong association between especially pragmatic skills and mental health in women in prison, but these are not simple associations, and there is probably reciprocal influence between a number of related factors. This is a situation that is well known in the field of language and related disorders, where a model including risk factors and protective factors is increasingly recognized as a meaningful frame of understanding (Bishop et al., 2017; Compton, 2021). Furthermore, we know that comorbidities are common, which again underscores the high degree of reciprocity in how different personal and environmental characteristics influence each other. This means that there should be an obvious place for professionals with knowledge of these factors in the justice system, e.g., speech and language pathologists and psychologists.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Norsk senter for forskningsdata (NSD). The patients/participants provided their written informed consent to participate in this study.

FM, WH, and LJ contributed to the conception and design of the study. EE and AV performed preliminary literature searches and analyses. FM performed statistical analyses and wrote the first draft of the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by a grant from the County Governor of Vestland, Norway.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.