Trauma, Rememory and Language in Holodomor Survivors’ Narratives Травма, спам’ятовування та використання мови у наративах свідків Голодомору

The objective of the research is to examine language use in Holodomor survivors ‘narratives as psycholinguistic markers of mental trauma and PTSD. The specific objective is to explore rememory as a cognitive strategy of releasing suppressed traumatic events. Materials & Methods. 42 survivors of the Holodomor of 1932–1933 in Ukraine were recruited for producing a traumatic narrative. The inclusion criterion for participants was their personal history of being Holodomor survivors. Holodomor survivor is defined as a person who was exposed to the genocide and unprecedented starvation in 1932–1933. The study took place in 2003–2005, average age of participants is 84.5, SD=4.8, 29 females and 13 males. The study applies LIWC (Linguistic Inventory Word Count) to analyze the traumatic narratives and captures linguistic units and the psychological meaningful categories. The study applies the exploratory design Trauma, Rememory and Language in Holodomor Survivors’ . . . 81 © Zasiekina Larysa utilizing the independent variables of categories of time, I and cognitive processes and dependent variable of word count in a traumatic narrative for multiple regression analysis, SPSS. 26. Results. The main issue that emerges from the findings is that categories of I, time, and cognitive processes taken together contribute to word count. However, only categories of time (positive predictor) and cognitive processes (negative predictor) are independent significant predictors of word count. Therefore, we can assume that a poor reappraisal of traumatic events and overestimation of time in the rememory of traumatic narratives indicate PTSD symptoms in Holodomor survivors. Conclusions. Rememory as a cognitive strategy has a positive impact on developing collective identity and filling gaps in the Ukrainian history, however, it does not affect the therapeutic effect in treating PTSD.


Introduction
Holodomor as a collective, historic and individual trauma is a major problem in multidisciplinary framework, in particular History, Psychology, Linguistics, Social and Political Sciences. Holodomor or Famine of 1932-1933 is associated with an increased risk of transgenerational transmission of trauma in offspring of this genocide in the Ukrainian population. Evidence consistently suggests that the direct descendants of Holodomor survivors have specifi c emotions (horror, fear, mistrust, sadness, ethnic-related shame, anger, stress and anxiety), and behavioral patterns, aligned with trauma (overeating, intrinsic selfi shness and indifference to others, overemphasis on food) (Bezo & Maggi, 2015). The trauma-based behavioral patterns relate to a strong fear of possible reoccurrence of hunger in future and expressed in seeking security linked to food (Chemtob et al., 1997: 184). Therefore, there is an urgent need to explore the mental trauma of Holodomor survivors, since it also addresses the safety and psychological wellbeing in offspring of the second and the third generations. Moreover, the enchancing emotions and inner states in the second generation of offspring in Holodomor survivors is aligned with healthy aging and an effi cient psychological support for this age group.
A number of cross-sectional studies suggest an association between genocide and mental trauma and moral injury in genocide survivors, traumatic memory reorganization and effect of traumatic narrative on effi cient treating post-traumatic stress disorder (PTSD) (Adenauer et al., 2011;Broom, Kfi r & Dasberg, 2001;Kira et al., 2008). Recently investigators have examined the effects of rememory on recollecting collective trauma and developing collective identity (Nikro, 2019). However, the linguistic mechanisms that underpin rememory as a cognitive strategy are not fully understood. Despite the importance of traumatic narrative in treating PTSD, there remains a paucity of evidence on language use in traumatic narratives and psycholinguistic markers of PTSD.
The objective of this study is to examine language use in Holodomor survivors 'narratives as psycholinguistic markers of mental trauma and PTSD. The specifi c objective is to explore rememory as a cognitive strategy of releasing suppressed traumatic events.
Traumatic memory is defi ned as traumatic neurosis and PTSD (an anxiety arising as a delayed and protracted response after experiencing or witnessing a traumatic event involving actual or threatened death or serious injury to self or others) in the Oxford dictionary (Colman, 2003). There are different kinds of mental trauma based on the kind of traumatic event (natural disasters, technological disasters, automobile accident) and on the specifi c victim population (combat veterans, rape victims, victims of domestic violence, victims of child sexual abuse, crime victims) (Meichenbaum, 1994). Evidence consistently suggests that there is a strong association between traumatic memory and PTSD, notably overgeneralization (Lorenzzoni, Silva, Poletto & Kristensen, 2008), fl ashbacks and intrusions (Conway, 2005), centrality of traumatic event (Bernsten, Rubin & Siegler, 2011), involuntary autobiographical memory (Rassmusen, Ramsgaard & Berntsen, 2015). Furthermore, PTSD as a mental disorder, aligned with alterations in individual memory, requires a recollection of traumatic memory. American Psychological Association revised the main criteria for PTSD, in particular exposure to a traumatic or stressful event, intrusion symptoms, negative alterations in cognition and mood, alterations in reactivity and arousal, duration (symptoms last more than one month), functional impairment and social exclusion. There also some specifi cations, notable depersonalization and derealization (APA, 2013). Trauma-focused cognitive therapy has been found to be an evidence-based treatment for PTSD (Chemtob, Novaco, Hamada & Gross, 1997).
Considering the specifi cations for PTSD, narrative-oriented therapies are suggested, in particular Trauma-Focused Cognitive and Behavioral Therapy, Narrative Exposure Therapy, which involve individuals into construction of their lifes, including emotions, cognitions, physiology, behavioural and sensory elements (Adenauer et al., 2011;Chemtob et al., 1997). Despite the importance of a traumatic narrative in PTSD treatment, there remains a paucity of evidence on rememory as a tool for producing a traumatic life event.
Rememory as a narrative device and a representational technique was introduced by Tony Morrison to defi ne the narrative strategy of subversive representation in articulating trauma, which it is impossible to narrate (Purkayastha, 2014). The concept rememory is reinvented to capture the recollection of a traumatic individual memory which is unspeakable because of its severe nature. Rememory is closely relates to Afro-American literature and expresses a traumatic experience of slavery. Unlikely the traumatic narrative, rememory is an opportunity for narrator to release individual suppressive traumatic episodes, and simultaneously to reconstruct the discrete and distorted episodes of traumatic experience in the complete and coherent narrative in historical context. Therefore, rememory is a bridge between psychological disintegration of a traumatized person at the individual level and collective traumatic memory, aligned with this trauma. Nikro (2019: 8) suggests to use term «dismemory» instead of «rememory» to defi ne a «more productive notion of forgetting and/or practice of memory as a convenient, politically expedient mode of historical amnesia -the sly collusion of amnesty and amnesia».
Since rememory is linked to the situations of unspeakable terror, grief, intolerable pain, the current study applies this concept to explore the traumatic experience of Holodomor survivors. As Kulchytskyi (2008) argues, the Holodomor of 1932-1933 was a terror by famine resulted at «the killing of people by creating conditions incompatible with life» (p. 90). In other words, it was a death sentence imposed on the Ukrainian population through an unbearable suffering and starvation. Therefore, rememory of Holodomor survivors at the individual level simultaneously represents the collective memory of this genocide in Ukrainians and revitalized historical amnesia. In other words, individual traumatic narratives of Holodomor survivors could be explored as a collective experience of Holodomor, fi lling the gaps in the Ukrainian Historiography. In the current research, rememory is termed as a complex process of releasing repressed traumatic memory at the individual and collective levels.
The recollection of collective traumatic memory aligned with the Famine of 1932-1933 is of crucial importance, since the factors leading to Holodomor were concealed by the Soviet government and even some governments of Ukraine as an independent state, therefore, still remain speculative. Thus the rememory of this tragedy at the individual and collective levels can integrate the Holodomor into a broader understanding of genocide.
Language as an individual ability to narrate is recognized as a main tool for traumatic narrative, which makes a sense for human experience (White, White, Wijaya & Epston, 1990). However, creating event-based traumatic narrative is not enough for reconstructing traumatic memory and enhancing individual mental health. It is important to express all thoughts and feelings towards the traumatic events and give them a meaningful experience (Pennebaker, 1993). The fi ndings from this research indicate that the improvement of health is associated with a higher frequency of words expressing negative emotions, insight, causal words, and cognitive processes. Thus these words serve as psycholinguistic markers of enhancing individual wellbeing and transforming rememory in a therapeutic narrative.
The fi ndings of recent research suggest that the words expressing external agent, place and time represent the propositional structure of traumatic memory, however time is a negative predictor of an individual's spontaneous self-expression in traumatic narrative of individuals without PTSD. The traumatic memory of individuals who experienced a traumatic event and successfully coped with it without PTSD, focuses on the external agent instead of the internal agent, and shifts focus from past, avoiding overestimation of time linked with the traumatic stressful event (Zasiekina, Kennison, Zasiekin & Khvorost, 2019).

Methods and Techniques of the Research
Considering the language units as psycholinguistic markers of traumatic memory and PTSD in Holodomor survivors, the present research tests the following hypothesis (H): H There is a linear relationship between pronoun I (as a psycholinguistic marker of depersonalization), time (as a marker of derealization), and cognitive processes (as a marker of reappraisal of traumatic experience) and word count (as a marker of spontaneous selfexpression) in the traumatic narrative.
Participants 42 survivors of the Holodomor of 1932-1933 in Ukraine were recruited for creating a traumatic narrative. The inclusion criterion for participants was their personal history of being Holodomor survivors. Holodomor survivor is defi ned as a person who was exposed to the genocide and unprecedented starvation in 1932-1933. The study took place in 2003-2005, average age of participants is 84.5, SD=4.8, 29 females and 13 males (see Table 1). The participants represented different oblasts in Ukraine and were recruited with the formal letters to take part in the study. After consent forms were obtained, the participants were asked to talk freely about the events with the major focus on their thoughts and feelings. After that the traumatic narratives were transcribed and analyzed.

Method
LIWC (Linguistic Inventory Word Count) analyzes the traumatic narratives and captures linguistic units and the psychological meaningful categories. LIWC as a software tool refers to the conceptual framework proposed by Tausczik and Pennabaker (2010) to select word meanings linked to the attentional focus, emotionality, social relationships, thinking styles, and individual differences. The current study applies LIWC to discover linguistic units and psychological meaningful categories aligned with PTSD symptoms, notably depersonalization (pronoun I), derealization (category of time, where past and present are mixed), cognitive processes (as a marker of reappraisal of traumatic experience). Since the traumatic narratives were collected and transcribed in Ukrainian, the study applies Ukrainian version of LIWC (Zasiekin et al., 2018). The study applies the exploratory design utilizing the independent variables of categories of time, I and cognitive processes and dependent variable of word count in a traumatic narrative for multiple regression analysis, SPSS. 26.  Table 2 illustrates the distribution of the main psychological categories in the traumatic narratives of the Holodomor survivors, the mean indices of distribution of the psychological categories in the autobiographical narrative (normal speech, suggested by Tausczik and Pennabaker) (2010) and traumatic narratives in Holodomor survivors, and Cronbach's alpha α to indicate the internal consistency for the personal pronoun I, and categories of time and cognitive processes. Multiple regression analysis has been used to predict the value of continuous variable (word count) based on other independent continuous variables, namely pronoun I, time, and cognitive processes. The assumptions of linear relationship, homoscedasticity, independence of residuals (Durbin Watson d=1.46), multicolleniarity (average Tolerance=0.98>0.1, and average VIF=1.06, thus 1<VIF>10) were met. Finally, the size of the sample is above number required by Central Limit Theorem (n=42>30), therefore the assumptions regarding normal distribution of data is also met.

Results
The results of the regression with a forced enter method show that three variables (I, time, and cognitive processes) explained 26% of word count, R 2 = .26, F (3,38) = 5.82, p = .002. The results also show that categories of time and cognitive processes signifi cantly predict the word count, respectively, b = .34, t (42) = 2.48, p = .020; b = -.42, t (42) = -3.03, p = .004. Pronoun I, b = .14, t (42) = .99, p = .32 does not signifi cantly predict the word count (see Table 3). The main issue that emerges from these fi ndings is that categories of I, time, and cognitive processes taken together contribute to word count. However, only categories of time and cognitive processes are independent signifi cant predictors of word count.
Since the points in a residual plot are randomly dispersed around the horizontal axis between 0-1.5, a linear regression model is appropriate for the data in the current research (see Figure 1).

Discussion
The study set out to examine language use in Holodomor survivors` narratives as psycholinguistic markers of mental trauma and PTSD.
With respect to the hypothesis, it was found that categories of time and cognitive processes are independent signifi cant predictors of word count. More specifi cally, the category of cognitive processes is negative predictor of word count. The possible explanation for this fi nding is that a low frequency of cognitive processes fi ts into PTSD symptoms, in particular avoidance and negative alterations in cognition. Therefore, a low frequency of cognitive processes in the traumatic narrative expresses a poor reappraisal of the traumatic events, which is aligned with PTSD. This fi nding corroborates with other studies indicating that improved health after PTSD treatment is associated with a higher frequency of words insight causal words, and cognitive processes (Pennebaker, 1993).
N.C. Hunt points out: «Recovery from trauma means making sense of it all again, learning to understand the world as it is in the light of traumatic event, incorporating the new trauma related information into one's narratives» (Hunt, 2010: 126).
Therefore, a low frequency of category cognitive processes is associated with the diffi culties to reappraise the traumatic events and accommodate it in the personal understanding of the world by the Holodomor survivors. This fi nding is in line with the data analysis by LIWC, notably a higher frequency of category cognitive processes in the autobiographical narratives comparatively with the traumatic narratives, respectively 12.91 and 9.10. The cognitive processes used in the traumatic narratives are mostly aligned with memory recollection, including individual thoughts and emotions towards the tragic events. However, rememory of Holodomor survivors is not accompanied with cognitive processes and expresses a poor reappraisal of the Holodomor, which results at poor psychotherapeutic effect at the individual level, however, contributes a lot to the developing of collective memory, e.g.

I remembered as NKVD pursued my father; I know and you know that it was a genocide; I have kept so many deaths in my memory for my entire life; I remember this hunger and will not forget it until the end of my life; I remember a lot of corpses at the Kramatorsk railway station.
As mentioned in the literature review, prior studies mostly have noted the importance of creating an individual traumatic narrative for treating PTSD. However, very little was found in the literature on the rememory as a strategy to release the suppressed memory not only at the individual, but also at the collective levels. In other words, rememory of these events leads to understanding the Holodomor as a collective trauma and developing collective identity in Ukrainians. However, rememory as a cognitive strategy is aligned with cognitive words of memory and mnemonic processes, and does not evolve the cognitive verbs of reappraisal (to be aware, to understand, to make a sense, to capture, to explore, to realize) as a necessary tool for making sense of the event in the entire life and recovering from PTSD. Therefore, rememory in traumatic narratives has a positive impact on developing collective memory and identity in Ukrainians. However, rememory does not affect the positive psychotherapeutic outcome in treating PTSD.
Consistent with the literature, this research found that participants, who witnessed the traumatic event and suffered PTSD, reported using a mixture of present and past time, escaping reality, named as derealization (Zasiekina et al., 2019). One unanticipated fi nding was that there is a higher frequency of category time in traumatic narrative comparatively with autobiographical narrative, respectively 4.28 and 2.17 (see Table 2). However, the study captures the highest discrepancy of category time focused on present in the autobiographical and traumatic narratives, respectively 15.28 and 1.74. One possible explanation might be that traumatic narrative is aligned with memory recollections of the past events, however, focus on the past could also related to PTSD symptoms, notably intrusions and fl ashbacks. This conclusion is confi rmed by the examples of the traumatic narratives, in particular the pictures of dead people from the past are always present in my mind; they arrested my father in summer 1933, I remember this day in all details; I think that the Holodomor was planned in 1929, after that time the terrifi ed images of hungry people pursued me.
Comparison of the fi ndings with those of other studies confi rms that time is a predictor of the length of traumatic narrative (Zasiekina et al., 2019;Pennebaker, 1993). However according to previous fi ndings time is a negative predictor of an individual's spontaneous self-expression in traumatic narrative in individuals without PTSD, since they do not overestimate time like individuals with PTSD (Zasiekina et al., 2019). Therefore, we can assume that focus on time, represented by a mixture of present and past, in the traumatic narratives indicate PTSD symptoms in Holodomor survivors.

Conclusions
The aim of the present research was to examine the language use in Holodomor survivors 'narratives as psycholinguistic markers of mental trauma and PTSD. This study has shown that that linguistic and psychological meaningful categories, notably personal pronoun I, cognitive processes and time taken together signifi cantly predict the length of traumatic narrative. The most obvious fi nding to emerge from this study is that time and cognitive processes as independent signifi cant predictors of word count in traumatic narratives indicate a high probability of PTSD in Holodomor survivors, aligned with a low reappraisal of these tragic events and a poor time orientation, resulted from a mixture of past and present, and overestimation of time.
One of the more signifi cant fi ndings to emerge from this study is that rememory as a cognitive strategy has a positive impact on developing collective identity and fi lling gaps in the Ukrainian history, however, it does not affect the therapeutic outcome in treating PTSD. The fi ndings of this research provide insights for a crucial importance of traumatic narrative incorporating thoughts and emotions towards a traumatic event instead of the focus on time and mnemonic processes in creating traumatic narrative while treating PTSD.
An implication of this is the possibility that Holodomor survivors suffer PTSD after a man-made Famine of 1932-1933. A limitation of this study is that it was not possible to assess PTSD with psychometric measures; therefore, it is only assumed that the Holodomor survivors suffered PTDS. However, this study has raised important questions about the nature of rememory and language use in Holodomor survivors in creating the traumatic narratives. These fi ndings have also signifi cant implications for the understanding of how rememory have an effect on developing collective memory and collective identity of Ukrainians. The results of the current study provide the insights for future research of language use in the traumatic narratives of a genocide survivors.