Дискурсивные изменения в публичной риторике президента Турции Эрдогана (2014–2025): корпусноориентированный количественный анализ

Раздел: Язык СМИ

В исследовании проведен количественный дискурс-анализ корпуса из 1888 публичных выступлений президента Турецкой Республики Реджепа Тайипа Эрдогана. Корпус охватывает период с 28 августа 2014 по 31 декабря 2025 г. и включает 3,3 млн словоупотреблений. Цель работы – проверить, каким образом «работа по легитимации» лидера в дискурсе реагирует на кризисные периоды, а также оценить, какие тематические линии сохраняют устойчивость во времени. Кризисный контекст включает попытку государственного переворота, пандемию и эпизоды экономической волатильности. Морфологические особенности турецкого языка потребовали дополнительной обработки, поэтому в анализе использованы морфологическое упрощение и консолидация понятий. Исследование отслеживает 48 ключевых концептов, отобранных на основе теоретически и эмпирически ориентированных процедур. Каждый концепт измеряется посредством нормализованной частотности на 1000 слов по годам. Лексическое разнообразие оценивается с помощью индекса Гиро, что снижает чувствительность показателя к длине текста. Полученные результаты противоречат распространенному тезису о том, что президентский дискурс со временем упрощается. Напротив, данные указывают на лексическое обогащение и институционализацию. Коллективно идентифицирующая и религиозно ориентированная лексика демонстрируют выраженную устойчивость в разные периоды. Данная закономерность особенно отчетливо проявляется в повторении «якорных» элементов. Лексика безопасности и терроризма заметно усиливается в 2016–2019 гг. После 2020 г. более значимыми становятся темы экономики и развития. В целом результаты показывают, что президентский дискурс организован посредством многослойной и адаптивной стратегии легитимации (outputbased legitimacy), центр тяжести которой изменяется в зависимости от контекста, а не следует единой идеологической оси.

DOI: 10.55959/msu.vestnik.journ.3.2026.2969

Introduction

Official speeches by heads of state are communicative texts that convey policy to the public in contemporary political systems. They also function as core discursive structures that reveal how power is established, maintained, and legitimized (Dunmire, 2012). When taken together, campaign speeches, parliamentary opening addresses, summit remarks, crisis statements, and routine policy announcements form an integrated leadership discourse. This discourse is continuous over time. It is layered across political moments. It is also ideologically directive (Kujanen, Koskimaa, Raunio, 2024). This continuity matters for more than what leaders say at particular moments. It helps scholars examine how leaders construct a discursive universe across contexts and across time. It also shows which themes are stabilized and which interpretive frames are reproduced (de Vreese, Esser, Aalberg, Reinemann et al., 2018).

Recent work on political discourse shows that leader language is not a passive mirror of political reality. It is a force that actively constructs reality. It also constrains and steers what becomes thinkable and legitimate (Wodak, 2020). Through speeches heads of state define social problems. They set priorities. They reproduce collective identities. They also delimit legitimate policy space through discourse (Bos and Brants, 2014). In this sense leader discourse is not only symbolic communication. It is also a governing technology that is tightly entangled with material practices of power, institutional decision making, and mechanisms of governance (Papathanassopoulos and Giannouli, 2025).

Scholarship on head-of-state discourse has expanded markedly in recent years. This growth is linked to intensified global uncertainty shaped by developments such as the pandemic, war, and the climate crisis. It is also linked to the increasingly personalized and mediatized character of political leadership (Messini, 2025). Contemporary research on populism, charismatic leadership, and performative politics shows growing centralization around the figure of the individual leader. It also shows that official presidential speeches play an increasingly visible and consequential role in building and sustaining political legitimacy (van Aelst, Sheafer, Stanyer, 2011). At the same time much of the literature focuses on discrete crisis moments. It also relies on selected speeches (Szalay and Rakovics, 2025; Fowobaje, Mashood, Ekholuenetale, Ibidoja, 2022; Kyriazi, 2026). Other studies examine short time windows (Sikanku, 2022; Motinyane, 2025). These approaches generate valuable insights. They still struggle to capture long-run dynamics of continuity and change in leader discourse. Yet presidential speech often builds a stable meaning system over time. It does so through recurring themes, durable metaphors, and strategic discursive shifts.

Political leadership is not only a process of administrative decision making. It is also a process of discursively framing those decisions to generate consent. In the twenty-first century expanded media environments and the personalization of politics have made leader rhetoric a constitutive element of political regimes. In Türkiye two institutional changes are especially consequential. In 2014 the president was elected by popular vote for the first time. In 2017 the constitutional amendment introduced the Presidential Government System. These shifts transformed the legitimacy basis of the executive. They also reshaped its communication strategy.

There is a substantial literature on Erdogan’s political career and rhetorical capacity. Most of this work relies on qualitative interpretation. It also tends to use limited samples of text. Quantitative research that tracks discursive change across the full presidential period remains rare. This gap is especially clear for studies that use a comprehensive corpus and large-scale methods.

Using a 3.3-million-word dataset that includes all public speeches from 2014 to 2025, this study addresses three core research questions.

1. How did the language of political legitimacy respond to macro crises such as the July 15 coup attempt, the pandemic, and economic volatility?

2. To what extent do processes described through securitization and populism theories align with the empirical patterns in the data?

3. Does the leader’s vocabulary simplify over time, or does it instead reflect institutionalization under the new system?

Conceptual Framework and Literature Review

In recent years the political field has been reconfigured around security discourses, populist styles of politics, and debates over sovereignty. At a global scale migration, the pandemic, terrorism, economic crises, and cultural identity conflicts are often framed as existential threats rather than technical or administrative problems. This framing enables political actors to move beyond the boundaries of ordinary politics (Kurylo, 2020). Recent research on migration shows that crises are increasingly recast as security issues. It also shows the tightening of border and surveillance regimes. These studies suggest that threat narratives have become decisive in policy design and in legitimation (Montagna, 2023). Work on the post-COVID-19 period points to a similar pattern. Populist actors reframe crisis and health risks through discourse. This line of research shows how security talk can operate as a performative style within populist politics (Lamour and Carls, 2022; Schenkel, 2024). In this context securitization theory offers a strong analytic framework. It explains how political issues are moved into an “extraordinary” domain through discursive practices (Kurylo, 2020).

Populism is also a defining feature of contemporary politics. It constructs society around a homogeneous “people” and an opposing “elite” or “other.” This antagonism is increasingly reinforced through securitycentered discourse (Wojczewski, 2020; Bonansinga and Kielty, 2025). Populist actors shape the agenda through rhetorical and performative strategies that dramatize threat perceptions. These discursive practices foreground insecurity and crisis narratives. They present decision making as urgent and exceptional intervention rather than ordinary politics (Magcamit and Arugay, 2024; Wojczewski, 2020). Recent studies emphasize migration, identity, and security politics. They show that populist discourse can reproduce a “vertical” people-elite divide. It can also reproduce a “horizontal” inside-outside opposition through securitization dynamics (Wojczewski, 2020; Bonansinga, Kielty, 2025). Populism should therefore be treated as more than an ideology or a style. It can also function as a mechanism that both produces and sustains securitizing practices.

The literature offers substantial theoretical contributions in securitization theory, populism research, and political theology. At the same time, relatively few studies integrate these approaches through a unified communication perspective that centers heads of state and their discursive practices. Securitization scholarship largely focuses on threat construction, speech acts, and the legitimation of exceptional measures (Balzacq, 2011; Stritzel, 2014). It pays less attention to how these processes operate through leader discourse, rhetorical style, and public meaning making. Populism research is often organized around peopleelite antagonism, crises of representation, and political style. It rarely specifies how this antagonism is communicatively built in presidential speech. It also underdevelops the narrative and affective strategies through which it is expressed (Moffitt, 2016).

Political discourse can be defined as a domain in which power relations are constructed, sustained, or transformed through language (Van Dijk, 1993). The theoretical backbone of this study draws on three approaches in political science. These are securitization theory, debates on populism, and political theology.

Crisis Governance and Securitization

Securitization theory associated with the Copenhagen School (Buzan, Wæver, de Wilde, 1998) describes a process in which a political issue is coded as an “existential threat” and removed from the sphere of normal politics. Once an issue is securitized the leader gains legitimacy to adopt extraordinary measures. The securitizing move gives the issue absolute priority over competing concerns. It also fosters the perception that other issues will lose significance unless the threat is addressed (Buzan, W ver, de Wilde, 1998). The securitizing actor is typically aware of this dynamic. The actor labels selected matters as security problems in order to justify exceptional action. Invoking “security” marks the issue as urgent. It also signals that all necessary measures should be taken to resolve the problem (O’Reilly, 2008, p. 69).

Populism

Ernesto Laclau (2018) defines populism not as a fixed and closed ideology. He treats it as a form of “political logic.” In this view “the people” or “nation” is not a pre-given and unchanging social subject. It is constituted through discourse. Through the mechanism of the “empty signifier” nation becomes a symbolic nodal point around which diverse and dispersed demands can be articulated. This approach provides a core theoretical basis for explaining how “nation” can be structured as a hegemonic identity in populist discourse. Cas Mudde and Cristóbal Rovira Kaltwasser (2017) define populism as a “thin-centered ideology.” For them the essence of populism is the opposition between nation and corrupt elites. Nation is portrayed as a morally superior and unified homogeneous community. Jan-Werner Müller (2016) defines populism through the claim to speak “in the name of the people” and through anti-elitism. He argues that the key feature of the populist leader is the claim to be the sole legitimate representative of the people.

Political Theology and Lexical Richness

Carl Schmitt (2005) argues that the major concepts of modern state theory are secularized theological concepts. This is the core claim of “political theology.” The sovereign’s decision-making position is supported through a theological form of legitimacy. Political theology does not treat leader discourse only as a communication tool. It situates discourse within normative and metaphysical foundations that make legitimacy claims possible. The approach foregrounds exceptional decision capacity in moments of crisis. It also highlights collective identity formation and the representation of authority. In contemporary scholarship political theology continues to matter for discourse analysis. Saba Mahmood (2016) and Talal Asad (2018) similarly emphasize how leader discourse reproduces legitimacy through religious and secular references. They also stress how these references are reworked in a post-secular public sphere. A common claim in the literature is that populist leader language becomes simpler and poorer over time (Moffitt, 2016). This study tests that claim in the Erdogan case using Guiraud Index evidence.

Research Design and Data Source

This study uses a quantitative content analysis design to examine longitudinal change in Recep Tayyip Erdogan’s presidential rhetoric. The population includes all public speeches delivered between August 28, 2014 and December 31, 2025.

The data were collected from the “Speeches” archive on the official website of the Presidency of the Republic of Türkiye (tccb.gov.tr). The corpus was compiled through web scraping and manual archiving. The final dataset includes 1888 standalone speech texts. It contains 3326661word tokens. It therefore constitutes a large corpus of presidential political discourse.

The table indicates a sharp increase from 2014 to 2015 across all three measures. Output remains high in 2016 despite a partial decline. A pronounced drop appears in 2017 for tokens, types, and speech count. From 2018 onward production rises again. The year 2019 stands out as a peak year for both total words and unique words. The period from 2020 through 2023 reflects consistently high output. Speech counts trend upward in this period and reach the maximum in 2023. That year also aligns with the presidential election cycle. After the election a clear decline occurs in 2024 across all measures. In 2025 output increases again. Levels do not yet return to the highs observed between 2019 and 2023.

Data Mining and Pre-processing

A multi-stage cleaning pipeline was implemented in Python to reduce statistical noise in the raw text and to convert the corpus into analyzable units.

1. Lexical normalization (case folding): All texts were converted to lowercase to prevent information loss caused by case sensitivity. This step avoids treating forms such as “Millet” and “millet” as different items.

2. Noise filtering (regex cleaning): Regular expressions were used to remove elements that can distort conceptual signals. These included statistical figures, budget numbers, date strings, URL addresses, and punctuation.

3. Stopword removal: Function words that structure Turkish grammar but carry limited standalone thematic meaning were removed. This group included conjunctions, particles, and pronouns such as “ve” (and), “ile” (with), “bu” (this), “şu” (that/this one), “ki” (that/which), and “için” (for) expressions that support rhetorical flow were also treated as contextual noise and excluded. Examples include “efendim” (well/sir/ma’am), “tabi” (of course), “sadece” (only/just), “var” (there is/exists), and “yok” (there isn’t/doesn’t exist).

Morphological Analysis and Stemming Strategy

Because Turkish is an agglutinative language, counting surface word forms can produce methodological bias. Standard stemmers can also introduce meaning loss in political terminology. To address these risks the study used a context-sensitive morphological simplification strategy paired with conceptual mapping.

Morphological reduction: Words were stripped of possessive, case, plural, and tense markers to recover a semantic base form. One example is the reduction of “ekonomimizdeki” (in our economy) to “ekonomi” (economy).

Conceptual consolidation (mapping): Variants within the same semantic cluster were merged under a single umbrella concept. For instance, ballot box (sandık) was used as an umbrella label for “sandıklar”, “sandıktan”, and “sandığa”. Likewise, election (seçim) was used as an umbrella label for “seçimler”, “seçime”, and “seçimden”. This approach allows the analysis to capture the concept’s rhetorical weight rather than dispersing it across inflected forms. Lemma or root generation relied on Zemberek vX (0.17.1), TRmorph vX (2025-11), rule-based regex and dictionary. Lemma outputs were manually checked. The most frequent error classes were proper name detection, homonymy, and abbreviations.

Sampling Strategy and Selection of Key Concepts

The 48 core concepts were selected through a mixed sampling strategy rather than random choice. The procedure had two stages.

1. Data-driven selection: A frequency scan produced a ranked list of the 2,000 most common roots in the corpus. This list captured the discourse’s baseline agenda.

2. Theory-driven selection: Concepts aligned with central debates in political science were then selected from the high-frequency list through purposive sampling. The guiding domains were securitization theory, populism, political economy, and identity politics. High-frequency verbs, such as do (yapmak) and come (gelmek), were excluded. Marker concepts, such as survival (beka), nation (millet), interest rate (faiz), and terrorism (terör), were retained.

This combined strategy ensured that selected items were both statistically salient and theoretically meaningful.

Statistical Normalization

The corpus volume varies substantially across years. For example, 2014 includes roughly 80,000 tokens while 2019 includes roughly 380,000 tokens. Using raw counts would inflate concept visibility in high-output years. The study therefore uses normalized frequency.

NormFreq = (n / N) × 1000

Here n is the count of the concept in a given year. N is the total number of tokens in that year. The resulting values indicate how many times a concept appears per 1,000 words. This enables valid year-to-year comparison.

Lexical Richness (Guiraud Index)

Yearly text length varies widely. The minimum is 79,987 tokens and the maximum is 380,506 tokens. This variability makes the type-token ratio unreliable because TTR tends to decrease as texts become longer. To reduce length sensitivity the study uses the Guiraud Index.

R = V / √N

V is the number of unique word types and N is the number of tokens. This measure provides a vocabulary richness indicator that is more comparable across years.

Data Validation and Limitations

To support reliability a random 5 percent of the processed data was manually validated. The estimated algorithmic error rate was below 0.5 percent. The findings should be interpreted within the following limitations. Semantic context: The analysis is frequency-centered. It does not model irony or negation. It also does not include sentiment analysis. Homonymy: Some items have meanings that vary by context. “Meclis” can refer to an institution or a general meeting. Fully automated disambiguation was not feasible. The study assumes that large corpus size reduces the effect of such noise on annual trends. This risk was further reduced through KWIC checks and sensitivity analysis.

Processing Infrastructure and Analytical Tools

Data mining, cleaning, and statistical analysis were conducted using a mixed toolchain. The primary environment was Python 3.15. A secondary tool was AntConc 4.3.1 for corpus validation.

Python-based Computational Analysis

A custom NLP pipeline was developed in Python to process the 3.3 million word dataset and to compute annual longitudinal measures. Data structuring (Pandas): The 1,888 text files were organized by year. They were converted into data frames. Yearly token volumes were computed. Pattern matching and cleaning (re): Dates, times, URLs, and numeric strings were removed. Stopword filtering was implemented using regular expressions. Frequency counting and normalization: Root counts and per-year NormFreq values were computed through custom functions. This automation reduced the risk of manual error.

Contextual validation with AntConc

AntConc 4.3.1 was used as a validation instrument to assess the linguistic plausibility of the Python outputs. The software is widely used in corpus linguistics and is associated with Laurence Anthony at Waseda University.

KWIC validation: For concepts with a higher homonym risk, the KWIC function was used to inspect contexts. Sudden frequency spikes in the Python results were cross-checked in AntConc using the concordance plot. A random sample of sentences was reviewed to confirm that each token matched the intended analytic sense. This dual-tool verification combines Python’s large-scale processing capacity with AntConc’s contextual precision. It increases confidence in the substantive validity of the quantitative measures.

Structural Change and Lexical Richness in The Discourse

Structural change in the leader’s discourse was examined through lexical richness and word diversity measures. Table 2 reports total tokens (N), unique types (V), the standard type-token ratio, and the Guiraud Index for the 2014–2025 period.

The results indicate a process of lexical adaptation and enrichment in Erdogan’s presidential rhetoric. This pattern runs counter to the common assumption that populist leaders’ language becomes simpler and less varied over time.

Analysis of the Findings

The TTR illusion and the Guiraud signal: Standard TTR declines from 22.8 percent in 2014 to 13.7 percent in 2015. It then rebounds to 16.1 percent in 2016. It rises sharply again in 2017 to 22.9 percent. Between 2019 and 2023 it remains in a narrow range from 13.3 to 14.4 percent. It increases in 2024 and 2025 to 16.4 percent and 15.8 percent. This pattern can be misleading because TTR is mechanically sensitive to corpus size. As token counts rise TTR tends to fall even if lexical resources expand. While the Guiraud Index reduces sensitivity to text length, a random downsampling procedure was applied to completely neutralize its residual mechanical reaction to extreme variations in speech volume. Each year’s corpus was equalized to the lowest annual token count (79,987 tokens in 2014) and the index was recalculated. Under this strict normalization, the index rises from a 2014 baseline of 64.45 to a new peak of 91.03 in 2018. By 2025, it remains highly elevated at 87.06.

This trajectory indicates that the discourse draws on a larger and more diverse lexical pool than in the early period.

System change and conceptual expansion: The enrichment trend from 2014 through 2018 aligns with Türkiye’s executive system transition. The shift is from a register shaped by day-to-day political contestation to a register oriented toward state building under the Presidential Government System. This institutional change increased the need to incorporate bureaucratic, military, diplomatic, and technical terminology into routine public address. The rise in lexical richness is consistent with an expanding conceptual repertoire. It suggests that the leader relied on a broader set of categories when articulating and legitimizing the new governing framework. Furthermore, because the corpus consists of pre-written, highly edited official texts rather than spontaneous colloquial speech, the language naturally resists the simplification typically associated with everyday populist rhetoric, reflecting instead the institutional weight of the state format.

Crucially, the sharp dip previously observed in the 2017 raw data is mathematically eliminated through downsampling (rising to 88.23). This proves that the apparent decline in lexical diversity during that year was merely an artifact of its lower token count, rather than a genuine linguistic simplification.

Stabilization and institutionalization: In the downsampled data, the Guiraud Index stabilizes in a higher 86 to 91 range in the subsequent years. This plateau suggests that the discourse reached a mature stage. It also indicates institutionalization of the core conceptual set. In practical terms the rhetoric appears to have consolidated a stable lexicon while maintaining high diversity rather than returning to a narrower or simplified vocabulary.

Collective Identity Discourse (Nation, Homeland, Martyr, Culture, Independence)

This cluster indicates a measurable shift in the legitimacy register of Erdogan’s rhetoric between 2014 and 2025. Rather than anchoring legitimacy primarily in a rational-legal framework, the discourse increasingly relies on a collective subject and shared values. It constructs what can be described as an “emotional community.” The quantitative pattern is consistent with a structural reorientation in the basis of legitimacy. The emphasis moves away from institutional belonging and toward existential belonging.

Across the eleven-year series, “nation” is the undisputed hegemonic center of the discourse. Its normalized frequency rises from 4.29 in 2014 to 6.78 in 2016 and to 6.96 in 2017. This is a dramatic increase of 62 percent. By 2025 the value remains high at 5.38. This level is 25 percent above the 2014 baseline and indicates stabilization. In political science terms this pattern is a direct empirical marker of populist legitimacy building through people-centrism (Mudde, 2004). The peak in 2017, during the referendum period, indicates that “national will” was coded as the primary source of legitimacy for the system change. The concept’s persistence at high frequency suggests that the presidency reproduces its legitimacy less through constitutional procedure and more through a claimed identification with the people themselves.

In 2014 “homeland” appears at 0.32. In 2016 it rises to 1.14. This is a record increase of 256 percent. It then stabilizes in the 0.48 to 0.75 range. This jump signals a territorialization of politics (Rossi, 2019). During the July 15 coup attempt and the trench operations period, political contestation shifts from routine struggles over power to the protection of territorial integrity and survival. As “homeland” becomes more central, the boundary between government and opposition is increasingly framed through the idiom of patriotism.

The term “martyr” follows a similar trajectory. It rises from 0.40 in 2014 to 1.46 in 2016. This is a 265 percent increase. It remains elevated and still reaches 1.06 in 2025. The surge and persistence indicate institutionalization of a sacrifice ethic within political rhetoric. Loss is framed less as mourning and more as sacred sacrifice. These framing positions affective legitimacy as a strategic resource for producing consent.

“Culture” increases sharply in 2017 and reaches 1.23. This represents a 162 percent rise. “Independence” also rises in 2017 to 0.27 and remains at roughly that level afterward. These patterns show that legitimacy is reinforced not only through security but also through a civilizational narrative organized around “local and national” values. The persistence of “independence” suggests that sovereignty is continuously framed through the analogy of a liberation struggle. It constructs political authority as a standing line of resistance.

From 2014 through 2016 all five collective identity markers show a clear upward trajectory. The increase culminates in 2016 and 2017. “Independence” and “nation” reach the highest levels in this phase. A synchronized decline appears across all items in 2018 and 2019. A limited rebound is visible in 2020. This is followed by another downturn in 2021. After 2022 the concepts move upward again. Between 2023 and 2025 “independence” and “nation” return toward upper-range values. “martyr” and “culture” increase in a more even and moderate manner. “Homeland” remains in a lower band than the other concepts. It still follows the same fluctuation pattern. Taken together Figure 2 indicates three phases in collective identity discourse. The first is a strong peak in 2016 and 2017. The second is weakening in 2018 and 2019. The third is renewed recovery after 2022.

Political Theology and Faith Language (Allah, God willing, praise be to God, faith, Islam)

This cluster shows that religious reference in Erdogan’s rhetoric is not an ornamental add-on that appears only when convenient. It operates as a structural backbone of the political register. In terms consistent with Carl Schmitt’s political theology, sovereignty and legitimacy are presented through theological reference that is blended with secular claims rather than derived from secular concepts alone.

The term “Allah” remains remarkably stable across the full period. It moves from 1.60 in 2014 to 1.67 in 2025 with minimal deviation. This stability persists despite major political, economic, and security shocks. The pattern indicates that the reference functions as a discourse constant. It helps position political authority on a transcendental plane. It also supports a form of spiritual legitimacy that is less vulnerable to worldly failure or crisis.

“God willing” follows a different trajectory. After dropping to 1.82 in 2016 it rises sharply after 2020. It reaches a peak of 3.34 in 2023. This pattern suggests that the language of theological hope is mobilized as a crisis governance resource. As uncertainty intensifies, including economic volatility and the pandemic, “god willing” becomes more salient than rational promise making. The term can therefore be read as a tool for managing public expectations through a metaphysical guarantee.

“Praise be to god” peaks in 2020 at 0.82. “Islam” rises in 2017 to 1.05 and then settles around 0.61 in subsequent years. The 2020 peak in “Praise be to god” aligns with the pandemic context. It indicates activation of a resilience register grounded in gratitude and endurance. The post-2017 routinization of “Islam” suggests mainstreaming. Religious politics appears less as a contested domain and more as a normalized component of official state discourse.

From 2014 through 2017 the series shows a general downward trend for “Islam,” “faith,” “praise be to God” and “Allah”. The year 2017 stands out as the low point for many items. From 2018 onward the pattern shifts upward. A pronounced increase appears in 2020. The series dips again in 2021 and then rises strongly. By 2023 almost all items reach peak levels. The highest values are concentrated in “Islam” and “faith.” “Allah” and “praise be to god” remain in a lower range. They still track the same cyclical movement. “God willing” stays more limited than the other items. It nonetheless moves in parallel with the broader periods of increase and decline. After a partial decrease in 2024 the series turns upward again in 2025. Overall, the figure indicates a strengthening of political theology and faith language after 2017. It reaches its highest intensity in 2023, which is again an election year.

Security, Terrorism, And Struggle Discourse (Terrorism, Security, Struggle, Organization, Operation, FETÖ, PKK)

This cluster provides quantitative evidence for a sharp securitization phase in 2016 through 2019. It also indicates institutionalization of an “emergency” register in state speech. The data suggest that security moves beyond a public service frame. It becomes a central parameter that defines the boundaries of the political field.

“Terrorism” rises sharply from 1.82 in 2015 to 4.06 in 2016. This is a 123 percent surge and it marks the historical peak. “Organization” follows a parallel pattern and reaches a record 3.38 in 2016. Both terms then decline steeply. By 2025 “terrorism” falls to 0.62 and “organization” to 0.36. Each end below its initial level. This exceptional concentration in 2016 captures the moment when an existential threat frame moved to the center of politics. The July 15 coup attempt¹ and the 2015–2016 trench clashes in southeastern Türkiye (hendek olayları)² pushed the discourse toward a friend-enemy logic that is consistent with Schmittian (2007) politics. The sharp decline by 2025 signals a confidence narrative. It suggests that the state presents the security crisis as overcome and the threat as marginalized. This pattern supports a shift from crisis governance to stability governance.

While label-based threat terms decline, “struggle” remains comparatively durable. After increasing in 2016 to 1.85 it retains much of its salience and still stands at 1.12 in 2025. This is substantively important. It indicates that even when named enemies fade from the foreground, the discourse sustains a standing “state of struggle.” Political legitimacy is kept active through vigilance and resistance frames. The object of struggle can change. It may be terrorism, inflation, or external powers. The struggle frame itself functions as an integrative political myth.

The abbreviation “FETÖ” appears for the first time in 2016. It is absent in the speeches before that point. It peaks in 2017 at 0.84. “PKK”³ (Kurdistan Workers’ Party) peaks in 2016 at 0.71. Both then recede to marginal levels by 2025. “FETÖ” drops to 0.20 and “PKK” to 0.00. These trajectories reflect shifting priorities within the state’s threat definition. The 2017 peak in “FETÖ” aligns with the post-coup purge and prosecution process and its legitimation needs. The subsequent attenuation of these labels strengthens the claim that threats have been neutralized or reduced to routine public order concerns.

From 2014 through 2016 security and struggle language shows a clear upward trajectory. The year 2016 is the peak across all items by a wide margin. “PKK,” “FETÖ,” “operation,” and “organization” reach their highest levels in 2016. “Struggle” and “security” also increase strongly in the same year. From 2017 onward the series begins a gradual decline across all concepts. The decrease continues through 2018 and 2019 and the curves settle into a more even pattern. After 2020 the downward movement persists and by 2021 the cluster stabilizes in a lower band. Between 2022 and 2024 the series shows limited fluctuation and the overall pattern remains flat at comparatively low levels. In 2025 a further decrease is visible across the full set. Taken together Figure 4 indicates an exceptional concentration of security and counterterrorism discourse in 2016. It then decreases in a stepwise manner and returns to a lower and more stable level in subsequent years.

Foreign Policy: Regionalization and Concretization (Syria, Cyprus, Azerbaijan)

This cluster documents a shift in foreign policy discourse. It moves away from abstract ideals associated with grand strategy. It moves toward concrete crises and bilateral partnerships tied to operational realities.

“Syria” remains in the approximate 1.0 to 1.5 range between 2014 and 2016. It then declines sharply and reaches 0.34 in 2021. This is a 78 percent drop. By 2025 it rebounds to 0.92 and appears more balanced. In 2016 Syria occupied the center of the discourse as a survival issue and an active war theater. The subsequent decline suggests institutionalization of border security, including the consolidation of “safe zone” practices. It also suggests routinization. The issue shifts from an urgent threat frame to a diplomatic and administrative file.

“Azerbaijan” reaches 0.34 in 2025 and “Cyprus” reaches 0.70. Both are the highest values in their respective series. This rise reflects a more concrete expression of Türkiye’s search for strategic autonomy. Azerbaijan’s victory in Nagorno-Karabakh in 2020 and the Eastern Mediterranean energy competition indicate a reorientation. The discourse shifts from a problem management frame centered on Syria to a gains-oriented and power projection frame centered on Azerbaijan and Cyprus. In quantitative terms the pattern supports a more proactive and revisionist tone in foreign policy rhetoric.

Between 2014 and 2016 the series for “Syria,” “Cyprus,” and “Azerbaijan” is volatile but remains comparatively elevated. In 2018 and 2019 all three items rise again. “Azerbaijan” and “Cyprus” move closer to upper-range levels in this period. After 2020 the cluster shifts downward. Between 2021 and 2023 all three items fall to their lowest bands. The sharpest decline appears in “Syria.” From 2024 onward the pattern turns upward again. In 2025 “Azerbaijan” reaches its highest level and “Cyprus” also increases strongly. “Syria” rises as well but remains more limited than the other two. Overall, the figure points to three phases in foreign policy discourse. It intensifies in 2018 and 2019. It weakens markedly in 2021 through 2023. It then strengthens again in 2024 and 2025.

Global System and Western References (World, Europe, NATO, Global)

This cluster indicates a shift in Türkiye’s positioning within the international system. The rhetoric moves from a Western-centered register toward a more global and critical register.

“World” remains elevated and fluctuates across 2014 through 2022. It falls to 2.29 in 2015 yet reaches a series high of 2.99 in 2022. A sharp decline appears in 2023 when the value drops to 2.03. The series then shows partial recovery in 2024 and 2025 and returns to the 2.28 to 2.37 range. Even as “world” declines, “global” increases, especially after 2020. This divergence indicates a shift in discursive substance rather than a retreat from international framing. During the pandemic and the conflict in Ukraine, the discourse increasingly focused on solving global crises and criticizing the international system. The slogan “The world is bigger than five” anchors this critique. Türkiye is framed less as a regional actor and more as a moral agent that claims authority to diagnose and address systemic dysfunction.

“Europe” declines steeply. It stands at 1.25 in 2016 during the refugee crisis period and falls to 0.39 in 2023. This is a 69 percent decrease. References to NATO also recede and reach a near-vanishing level by 2025 at 0.04. These patterns suggest that Western institutions lose reference value in foreign policy rhetoric. European Union membership or NATO alliance identity no longer operates as a routine target or anchor. Instead, it appears mainly as a technical file activated during episodic disputes. The Finland and Sweden accession debates illustrate this shift. Overall, the discursive decoupling appears deeper and more durable than any formal institutional break.

Between 2014 and 2016 “global,” “Europe,” NATO, and “world” follow a volatile but comparatively elevated pattern. From 2017 through 2019 the series shows modest decline and largely horizontal movement. Beginning in 2020 the pattern turns upward again. In 2021 and 2022 “global” reaches its highest level in the series. “Europe” and “NATO” also show a clear rise in 2022. In 2023 all items drop sharply. The decline is especially pronounced for “global” and NATO. In 2024 and 2025 the series recovers stepwise. Values remain below the 2022 peak. Overall, the figure indicates concentration of global system and Western reference language in 2022, strong weakening in 2023, and limited recovery afterward.

Institutional Politics and Regime Language (Politics, Parliament, Law, Constitution, Democracy, Freedom, Republic)

This cluster captures one of the clearest post-2017 shifts in Erdogan’s rhetoric. It documents a process of discursive de-institutionalization after the 2017 constitutional change. The data suggest that legitimacy moves away from procedural and constitutional language. It is increasingly built through leadership performance and collective identity.

“Politics” declines sharply from 2.34 in 2014 to 1.10 in 2016. This is a 53 percent drop. It then remains in a lower band. “Parliament” rises to 0.91 in 2017, which aligns with referendum-driven debate over institutional change. It then falls and reaches 0.39 in 2025. This is a 57 percent decline and it is among the lowest values in the series. The pattern does not imply an end of politics or a post-political condition. It points to a change in how politics is organized and narrated. The shift from a parliamentary system to a presidential system reduces the discursive weight of parliament. The rhetoric moves away from deliberative politics language and toward an executive action register.

One of the steepest declines appears in “democracy.” The value falls from 0.88 in 2014 to 0.21 in 2025. This is a 76 percent decrease. “Constitution” peaks only in 2017 at 0.67 during the referendum cycle and remains marginal in other years around 0.11. The sharp decline in democracy talk suggests a functional repositioning of the concept. It appears less as a stable normative commitment and more as a situational resource invoked in specific contexts such as elections or anti-coup framing. After 2016 legitimacy is increasingly articulated through “national will” and survival frames. In this configuration liberal-democratic norms such as freedom or separation of powers lose rhetorical centrality. Democracy language therefore becomes less operational in routine legitimation.

“Republic” declines from 1.09 in 2014 to 0.75 in 2025. “Law” declines more modestly from 0.70 to 0.60. These concepts are not abandoned. They do appear to lose their role as constitutive anchors. In the discourse “republic” is more often activated through commemorative and historical milestones such as the centennial rather than through sustained debate over regime principles.

In 2014 all items associated with institutional politics and regime language appear at relatively high levels. In 2015 and 2016 the series declines across the full set. This includes “republic,” “democracy,” “freedom,” “law,” “constitution,” “parliament,” and “politics.” A rebound appears in 2017. The increase is most visible for “republic.” Between 2018 and 2021 the pattern is more volatile yet it remains lower and more even overall. The year 2021 marks a relatively low point for many items. From 2022 through 2024 the series recovers again. “Republic,” “democracy,” and “freedom” rise in a stepwise manner. In 2025 a new decline is visible across all items. Taken together the figure indicates that institutional politics and regime language is strong in 2014, weakens in 2015 and 2016, recovers in 2022 through 2024, and then declines again in 2025.

Economic Discourse and Development Language (Economy, Investment, Production, Employment, Development, Budget, Inflation, Interest Rate)

This cluster becomes the new center of gravity after 2020. Fritz W. Scharpf’s concept of output legitimacy provides a strong theoretical lens for this period (Scharpf, 1999). As security threats recede, the discourse increasingly reconstructs legitimacy through economic performance and service delivery.

“Economy” reaches its peak in 2022 at 1.31. The more striking movement appears in “production” and “employment.” “Production” increases roughly fourfold relative to 2014. “Employment” rises even more sharply and peaks in 2021 at roughly six times the 2014 level. Together these patterns document the rhetorical construction of the “Türkiye Economy Model.” Rather than foregrounding abstract financial indicators such as interest rates or the stock market, the discourse shifts toward concrete outputs linked to everyday life. It emphasizes jobs, income, and factories. The rising salience of “employment” and “production” functions as a balancing strategy. It counters dissatisfaction generated by economic distress through a growth and development narrative.

“Inflation” is negligible for much of the series and then rises to a peak of 0.43 in 2022. In political rhetoric a problem tends to be named more frequently once it exceeds manageable boundaries. The spike therefore suggests that inflation became difficult to bracket or deflect. It also suggests that the leadership had to recognize it as a primary agenda item. The discourse often frames inflation as the product of external shocks and global crises. It then places the issue within a register of national economic struggle.

“Budget” starts at a very high level in 2014 at 2.08 and falls to 0.79 by 2025. References to “interest rate” nearly disappear in 2024. The decline in budget talk suggests a rhetorical exit from frames associated with fiscal discipline and IMF-indexed policy orientation. Legitimacy is no longer built through balanced-budget language. It is increasingly built through the promise of large investments.

“Interest rate” peaks in 2022. “Inflation” reaches its highest levels in the 2021 to 2022 window. At the same time “production” and “employment,” which remain relatively muted after 2016, rise rapidly after 2020 and reach a clear peak in 2021 and 2022. This pattern indicates a shift in the center of gravity of economic rhetoric. The discourse moves toward concrete domains of economic activity and it amplifies the production and employment frame. As production, employment, and development become more salient, the economic narrative is built around categories that carry more visible social referents.

“Development” and “budget” also increase strongly after 2020 and peak in 2022. This pattern shows that the economic frame is supported not only through macro indicators but also through an emphasis on growth and public resources. The year 2022 stands out as the point when nearly all items reach their highest levels at the same time. From 2023 onward the series declines across the full set. This includes interest rates, inflation, budget, development, production, and employment. In 2024 and 2025 the pattern becomes more even but it remains below the earlier peak.

“History” Discourse

This cluster captures the temporal dimension of political discourse. It shows how the past is mobilized to legitimize the present.

Among the 48 concepts tracked in this study, “history” is one of the few items that shows stable and near-linear growth independent of crisis cycles. Its normalized frequency rises from 1.08 in 2014 to 1.77 in 2025. This is a 64 percent increase. The pattern supports the claim that Erdogan’s political style is distinctly historicist. Contemporary political struggles, including episodes such as the Eastern Mediterranean disputes or counterterrorism campaigns, are framed less through present-day constraints and more as continuations of historical confrontation. References such as “Crusader versus Crescent” and “Sèvres versus Lausanne” function as templates for this coding. In this register history is not treated as an academic discipline. It operates as a strategic resource drawn from collective memory and used to structure and dominate present politics.

Taken together the longitudinal analysis across 2014 through 2025 indicates that presidential discourse is not one-dimensional. It cannot be reduced to a single Islamist or nationalist axis. Instead, the results point to a dynamic, adaptive, and multilayered legitimacy framework.

The “history” series rises stepwise from 2014 through 2016 and reaches its first clear local peak in 2016. A modest decline appears in 2017. The series then increases again between 2018 and 2020 and remains in a relatively high and stable band. A slight decline is visible in 2021. The series rises again in 2022. A small decrease follows in 2023. The pattern turns upward again in 2024 and 2025. The highest level is reached in 2025. Overall, history discourse shows limited volatility. It follows a steady trajectory with mild growth over time.

Taken together, the transformation of discursive hegemony can be summarized in three phases.

Securitized legitimacy (2015–2019): This phase is marked by peaks in terrorism, homeland, martyr, and survival language. Politics is organized around a friend-enemy axis. It is framed through existential threat.

Identity and value legitimacy (structural continuity): Nation, God, and history language forms the discursive backbone. These anchors remain resilient across crises. The leader-public relationship is reinforced through cultural and religious codes.

Performance legitimacy (2020–2025): As named security threats recede, legitimacy work shifts toward delivery and outcomes. Economy, investment, production, and global vision language becomes more central. This phase is framed as governance through action and results.

Overall, the evidence indicates that procedural democracy and institutional oversight language, including parliament and constitution references, weakens in the rhetoric. At the same time, legitimacy is increasingly organized around a plebiscitary leadership model. It rests on a three-part foundation. The foundation is historical mission, national unity, and economic development. The quantitative patterns support this conclusion with strong consistency across measures.

Discussion and Conclusion

This study uses quantitative evidence to trace structural change in Recep Tayyip Erdogan’s political discourse across an eleven-year period (2014–2025), spanning the institutional transition to the Presidential Government System. The central finding is that this discourse cannot be reduced to a single ideological line – neither “simply Islamist” nor “simply nationalist.” Instead, it operates as a strategically adaptive, multi-register legitimation repertoire: it reweights its emphases as challenges and opportunity structures change, while retaining stable symbolic anchors.

Existing research on heads-of-state discourse tends to cluster around three broad themes: (i) the structural organization of persuasion strategies, (ii) the role of context in shaping rhetorical choices, and (iii) the capacity of leader discourse to generate and sustain legitimacy. Corpusbased comparative work by Abd Aljabar Alkhlel, Mirsaeedi, Mohammed Abbas Alkhateeb, Vaez Dalili (2024), for example, contrasts rhetorical strategies in U.S. presidential speech with European leader rhetoric. Their findings point to shared persuasive repertoires but also to systematic differences in emphasis, suggesting that leader rhetoric is patterned by political and cultural context rather than individual preference alone.

A similar conclusion emerges in Ghasemi (2020), which compares Barack Obama and Hassan Rouhani. While both leaders rely heavily on logos-oriented argumentation, the study suggests that legitimacy-building draws on different reservoirs: universalistic value language in Obama’s case, and cultural-historical reference in Rouhani’s. In the security domain, Raissouni (2020) argues that “war on terror” discourse is not merely descriptive but constitutive: it can shape public perception through moral and rational foundations that legitimate policy choices. Complementing this line, Al-Khawaldeh, Rababah, Khawaldeh, Banikalef (2023) analyze Joe Biden’s inaugural address and emphasize the deliberate construction of strategies such as metaphor, contrast, reference, and intertextuality, designed to operate simultaneously on affective and logical registers. Taken together, these studies support an understanding of leader rhetoric as planned, layered, and context-sensitive.

Work on particular speeches similarly highlights how discourse is engineered to build consent and normalize policy frames. Salman Hummadi, Said, Manan (2019), analyzing Donald Trump’s Jerusalem speech through a motivated sequence approach, argue that religious and political references are structurally organized to secure audience alignment. Moving from case studies to large-scale regularities, Ficcadenti, Cerqueti, Ausloos (2019) use text-mining to identify recurring lexical patterns and structural templates in U.S. presidential speeches, interpreting these as evidence of institutionalized rhetorical conventions that reproduce long-run norms and shared values. This broader corpus perspective implies that leader discourse is not ad hoc; it is produced within recognizable constraints and rule-like patterns.

Recent work also underscores the interplay between crisis narration and agenda narration. Hamed and Alqurashi (2025), comparing Trump and Kamala Harris, distinguish conflict-centered crisis narration from agenda-centered vision narration and link these differences to political positioning. Mohammed (2025) likewise argues that effective leader discourse relies on recurrent tools such as repetition, metaphor, narrative, and inclusive language, with adaptation to historical context and audience expectations as a primary driver of rhetorical success. Finally, Nasakina, Kolisnichenko, Rohalska-Yakubova, Chepelyuk (2020), comparing leaders from the U.S., the United Kingdom, and Ukraine, identify metaphor, repetition, and emphasis as broadly shared persuasive devices across contexts. The implication is that leader discourse is culturally specific in content, yet often trans-contextual in its persuasive mechanics.

Against this literature, the present study adds a longitudinal, normalized account of discursive reweighting in a single leader corpus across major political and security-economic turning points. Before turning to interpretation, it is important to foreground several high-magnitude shifts that make the trajectory empirically visible. First, lexical richness increases rather than simplifies: when normalized through strict downsampling to control for text length, the Guiraud Index (a lexical diversity proxy) rises from 64.45 in 2014 to a peak of 91.03 in 2018 and remains high at 87.06 in 2025. Second, the securitized lexicon concentrates sharply around 2016: “terrorism” rises from 1.82 (2015) to 4.06 (2016), while “homeland” increases from 0.32 (2014) to 1.14 (2016) and “martyr” from 0.40 (2014) to 1.46 (2016). Third, people-centrism is structurally persistent: “nation” rises from 4.29 (2014) to 6.96 (2017) and remains elevated at 5.38 in 2025. Finally, procedural-institutional language weakens markedly across the series: “democracy” falls from 0.88 (2014) to 0.21 (2025), and “parliament” declines from its 2017 peak (0.91) to 0.39 in 2025. These are not marginal movements; they constitute the quantitative backbone of the argument.

On this basis, the findings support three substantive conclusions. First, the 2015–2019 period aligns with core claims in securitization theory. The exceptional concentration of counterterrorism and survival markers in 2016 indicates a shift into existential crisis governance following the July 15, 2016 coup attempt and the 2015–2016 “trench” conflict episodes. Legitimacy is constructed through protective leadership, in which threat narration and unity narration become mutually reinforcing. The subsequent decline of named-enemy markers after 2020 suggests that securitization here functions as a contingent legitimacy strategy rather than a permanently dominant regime feature.

Second, the post-2020 rise of economy, production, and investment language indicates a shift toward output-based legitimacy. The 2020– 2022 interval is the period when economic discourse becomes hegemonic: synchronized increases in interest rate, inflation, budget, development, production, and employment references show that the economic frame gains discursive priority. In the same interval, global system and Western reference language intensifies most clearly in 2022, suggesting that economic contention is narrated through an international positioning register (global system, global vision, and episodic disputes). In this configuration, consent is stabilized through delivery, performance, and development rhetoric rather than through continuous emergency governance.

Third, and most consistently, nation, God, and history operate as high-frequency anchors across the full series. These discourse constants remain resilient across terrorism, pandemic, and economic cycles. The persistence of “nation” at elevated levels after the 2017 referendum period, together with the steady rise of historicist coding (history increasing from 1.08 in 2014 to 1.77 in 2025), indicates that legitimacy is repeatedly grounded in collective identity, theological resonance, and historical mission claims. In practice, these anchors provide continuity when the center of gravity shifts from securitized to performance-oriented legitimation. The concurrent weakening of procedural-democratic and parliamentary vocabulary is also consistent with a rhetorical re-centering that accompanies institutional reconfiguration under the Presidential Government System.

Taken together, the 3.3-million-word corpus does not support a model of a single ideological axis or a linear trajectory. Instead, it supports a cyclical model of discursive reweighting: security peaks in 2016, economic performance becomes dominant in 2020–2022, and global reference language concentrates in 2022. At the same time, lexical richness rises and then stabilizes at a high level, running against the common “simplification” claim about populist leader language. The broader implication is that Erdogan’s rhetorical strength lies less in maintaining one consistent frame and more in mobilizing an adaptive legitimacy framework – security-centered in acute crisis, developmentalist under economic stress, and nationally anchored across periods – while gradually reducing the centrality of procedural-democratic and institutional oversight vocabulary.

This study contributes to the political communication and discourse analysis literature by demonstrating, with longitudinal and normalized measures, that legitimacy work can be both structurally anchored and contextually elastic. It also helps connect two literatures that are often treated separately: securitization scholarship focused on crisis peaks, and populism scholarship emphasizing durable people-centrism. The data suggest that these operate as an integrated repertoire rather than as mutually exclusive rhetorical regimes.

Примечания

¹ В ночь на 15 июля 2016 г. группа военнослужащих Вооруженных сил Турции предприняла попытку свержения демократически избранного правительства. Согласно официальной версии Министерства иностранных дел Турецкой Республики, путчисты использовали танки, истребители и вертолеты, пытаясь захватить ключевые объекты в Анкаре и Стамбуле. В Анкаре под удар попало несколько государственных учреждений, включая Великое национальное собрание Турции и президентский комплекс. Официальный нарратив приписывает попытку Террористической организации фетхуллахистов (FETÖ). Сообщается о 251 погибшем и более чем 2000 раненых среди мирного населения. Попытка провалилась благодаря антипутчистской позиции значительной части командования, действиям сил безопасности и массовому гражданскому сопротивлению. См.: Çavuşoğlu M. (2021) “Why Should We Fight Against FETO Resolutely?” Republic of Türkiye Ministry of Foreign Affairs. Available at: https://www.mfa.gov.tr/sayin-bakanimizin-neden-feto-ile-kararlilikla-mucadele-etmeliyiz-baslikli-mak... (accessed: 03.01.2026); Çavuşoğlu M. (2022) “Terrorism has no nationality, ethnicity nor religion. FETO threatens humanity as a whole.” Republic of Türkiye Ministry of Foreign Affairs. Available at: https://www.mfa.gov.tr/sayin-bakanimizin-terorun-milliyeti-etnik-kokeni-veya-dini-yoktur--feto-tum-i... (accessed: 03.01.2026); Statement On The Foiled Coup Attempt. Turkish Consulate General in Jerusalem. 17.07.2016. Available at: https://jerusalem-cg.mfa.gov.tr/Mission/ShowAnnouncement/308836 (accessed: 03.01.2026).

² Со второй половины 2015 по 2016 г. в некоторых городах на юго-востоке Турции происходили интенсивные столкновения, известные в обществе как hendek olayları (события вокруг траншей). В решениях, обобщенных Конституционным судом Турции, РПК описывается как партия, роющая траншеи на улицах, возводящая баррикады и минирующая их, т. е. пытающаяся установить фактический контроль в отдельных районах под лозунгом «самоуправления». Министерство внутренних дел Турецкой Республики и другие органы власти сообщают, что в отдельных населенных пунктах вводился комендантский час для защиты гражданских лиц и восстановления порядка и что проводились операции по устранению траншей и баррикад. Подчеркивается, что операции длились месяцами; сообщается о значительных потерях среди сил безопасности и уничтожении большого количества взрывчатых веществ. Агрегированная статистика жертв в связи с атаками начиная с 22 июля 2015 г. доводилась до общественности через заявления министерств. См.: Constitutional Court of the Republic of Türkiye (Anayasa Mahkemesi), decision of 23 May 2018, individual application no. 2016/29411. Available at: https://kararlarbilgibankasi.anayasa.gov.tr/BB/2016/29411 (accessed: 03.01.2026); Constitutional Court of the Republic of Türkiye, decision of 14 December 2023, individual application no. 2019/38701. Available at: https://kararlarbilgibankasi.anayasa.gov.tr/BB/2019/38701 (accessed: 03.01.2026).

³ В правовой и политической системе безопасности Турции РПК официально признана террористической организацией и рассматривается как один из главных источников политического насилия и внутренней угрозы безопасности для турецкого государства.

Библиография

Abd Aljabar Alkhlel R. M., Mirsaeedi A., Mohammed Abbas Alkhateeb M., Vaez Dalili M. (2024) Power of Persuasion: Analysis of U.S. Presidents’ and European Leaders’ Discourse. Journal of New Trends in English Language Learning (JNTELL) 3 (3).

Al-Khawaldeh N. N., Rababah L. M., Khawaldeh A. F., Banikalef A. A. (2023) The Art of Rhetoric: Persuasive Strategies in Biden’s Inauguration Speech: A Critical Discourse Analysis. Humanities and Social Sciences Communications 10: 936. DOI: 10.1057/s41599-023-02450-y.

Asad T. (2018) Secular Translations: Nation-State, Modern Self, and Calculative Reason. New York: Columbia University Press.

Balzacq T. (2011) Securitization Theory: How Security Problems Emerge and Dissolve. London: Routledge.

Bonansinga D., Kielty B. (2025) The Role of Insecurity in Left Populist Ideology: A Review of Podemos. Political Studies Review 23 (4): 1083–1093. DOI: 10.1177/14789299251326156.

Bos L., Brants K. (2014) Populist Rhetoric in Politics and Media: A Longitudinal Study of the Netherlands. European Journal of Communication 29 (6): 703–719. DOI: 10.1177/0267323114545709.

Buzan B., Wæver O., de Wilde J. (1998) Security: A New Framework for Analysis. Boulder: Lynne Rienner Publishers.

De Vreese C. H., Esser F., Aalberg T., Reinemann C., Stanyer J. (2018) Populism as an Expression of Political Communication Content and Style: A New Perspective. The International Journal of Press/Politics 23 (4): 423–438. DOI: 10.1177/1940161218790035.

Dunmire P. L. (2012) Political Discourse Analysis: Exploring the Language of Politics and the Politics of Language. Language and Linguistics Compass 6 (11): 735–751. DOI: 10.1002/lnc3.365.

Ficcadenti V., Cerqueti R., Ausloos M. (2019) A Joint Text Mining-Rank Size Investigation of the Rhetoric Structures of the U.S. Presidents’ Speeches. Expert Systems with Applications 123: 127–142. DOI: 10.1016/j.eswa.2018. 12.049.

Fowobaje K. R., Mashood L. O., Ekholuenetale M., Ibidoja O. J. (2022) Qualitative Content Analysis of Nigerian Heads-of-State and Presidents’ Inaugural Addresses: Text Mining, Topic Modelling and Sentiment Analysis. SN Social Sciences 2 (12): 279. DOI: 10.1007/s43545-022-00570-x.

Ghasemi F. (2020) Persuasive Language in Presidential Speeches: A Contrastive Study Based on Aristotelian Rhetoric. Buckingham Journal of Language and Linguistics 12: 19–38. DOI: 10.5750/bjll.v12i.1872.

Hamed D. M., Alqurashi N. (2025) Impact vs. Vision: A Critical Discourse Analysis of Trump and Harris’ Leadership Rhetoric in the 2024 Presidential Election. Frontiers in Communication 10: 1541513. DOI: 10.3389/fcomm.2025.1541513.

Kujanen M., Koskimaa V., Raunio T. (2024) President’s Constitutional Powers and Public Activism: A Focused Analysis of Presidential Speeches under Finland’s Two Presidencies. Comparative European Politics 22 (5): 594–615. DOI: 10.1057/s41295-023-00375-z.

Kurylo B. (2020) The Discourse and Aesthetics of Populism as Securitisation Style. International Relations 36 (1): 127–147. DOI: 10.1177/0047117 820973071.

Kyriazi A. (2026) Framing the EU Polity: How Commission Presidents Address Crises and Shape the Union. Journal of European Integration 48 (1): 23–43. DOI: 10.1080/07036337.2025.2476472.

Laclau E. (2018) On Populist Reason. Reprint edition. London: Verso.

Lamour C., Carls P. (2022) When COVID-19 Circulates in Right-Wing Populist Discourse: the Contribution of a Global Crisis to European Meta-Populism at the Cross-Border Regional Scale. Journal of Contemporary European Studies: 1–14. DOI: 10.1080/14782804.2022.2051001.

Magcamit M. I., Arugay A. A. (2024) Explaining Populist Securitization and Rodrigo Duterte’s Anti-Establishment Philippine Foreign Policy. International Affairs 100 (5): 1877–1897. DOI: 10.1093/ia/iiad248.

Mahmood S. (2016) Religious Difference in a Secular Age: A Minority Report. Princeton: Princeton University Press.

Messini S. (2025) Populist Political Communication on Instagram in the 2024 European Elections. Media and Communication 13: 10614. DOI: 10.17645/mac.10614.

Moffitt B. (2016) The Global Rise of Populism: Performance, Political Style, and Representation. Stanford: Stanford University Press. DOI: 10.2307/j.ctvqsdsd8.

Mohammed E. O. (2025) Persuasive Rhetoric in Public Speaking: A Discourse Analysis of Influential Leaders. International Journal of Research in English 7 (1): 598–611. DOI: 10.33545/26648717.2025.v7.i1j.395.

Montagna N. (2023) Quarantine Ships as Spaces of Bordering: The Securitization of Migration Policy in Italy during the COVID-19 Pandemic. International Migration Review 58 (2): 499–521. DOI: 10.1177/01979183 231154560.

Motinyane M. (2025) Performing the State: Continuity and Change in Ramaphosa’s Discursive Construction of Nationhood (2018–2025). Cogent Arts & Humanities 12 (1): 2584411. DOI: 10.1080/23311983.2025.2584411.

Mudde C. (2004) The Populist Zeitgeist. Government and Opposition 39 (4): 541–563. DOI: 10.1111/j.1477-7053.2004.00135.x.

Mudde C., Kaltwasser C. R. (2017) Populism: A Very Short Introduction. Oxford: Oxford University Press.

Müller J.-W. (2016) What Is Populism? Philadelphia: University of Pennsylvania Press.

Nasakina S. V., Kolisnichenko N., Rohalska-Yakubova I. I., Chepelyuk N. I. (2020) Rhetoric Strategies in the Presidents’ Speeches: Ukrainian versus English. Arab World English Journal (AWEJ) Special Issue on English in Ukrainian Context: 330–341. DOI: 10.24093/awej/elt3.27.

O’Reilly C. (2008) Primetime Patriotism: News Media and the Securitization of Iraq. Journal of Politics and Law 1 (3): 66–72. DOI: 10.5539/jpl.v1n3p66.

Papathanassopoulos S., Giannouli I. (2025) Political Communication in the Age of Platforms. Encyclopedia 5 (2): 77. DOI: 10.3390/encyclopedia5020077.

Raissouni I. (2020) The Language of American Political Discourse: Aristotle’s Rhetorical Appeals as Manifested in Bush’s and Obama’s Speeches on the War on Terror. International Journal of Linguistics, Literature and Culture 6 (4): 38–48. DOI: 10.21744/ijllc.v6n4.904.

Rossi F. M. (2019) Conceptualising and Tracing the Increased Territorialisation of Politics: Insights from Argentina. Third World Quarterly 40 (4): 815–837. DOI: 10.1080/01436597.2018.1465815.

Salman Hummadi A., Binti M. Said S., Bin A. Manan A. (2019) Persuasive Organizational Patterns and Rhetorical Arguments in Donald Trump’s Policy Speech on Jerusalem. International Journal of Psychosocial Rehabilitation 23 (2): 964–982.

Scharpf F. W. (1999) Governing in Europe: Effective and Democratic? Oxford: Oxford University Press.

Schenkel M. (2024) Health Emergencies, Science Contrarianism and Populism: A Scoping Review. Social Science & Medicine 346: 116691. DOI: 10. 1016/j.socscimed.2024.116691.

Schmitt C. (2005) Political Theology: Four Chapters on the Concept of Sovereignty. Chicago: University of Chicago Press.

Schmitt C. (2007) The Concept of the Political: Expanded Edition. Chicago: University of Chicago Press.

Sikanku G. E. (2022) Presidential Discourse, the Public and Recurring Themes: A Political Communication Analysis of the 2019 State of the Nation Address in Ghana. Communication and the Public 7 (4): 176–187. DOI: 10.1177/205 70473221129652.

Stritzel H. (2014) Security in Translation: Securitization Theory and the Localization of Threat. London: Palgrave Macmillan.

Szalay A., Rakovics Z. (2025) Tuned to Fear: Analyzing Viktor Orban’s State of the Nation Addresses, Focusing on the Enemy Images Identified in the National Consultation. Intersections. East European Journal of Society and Politics 11 (1): 165–184. DOI: 10.17356/ieejsp.v11i1.1226.

Van Aelst P., Sheafer T., Stanyer J. (2011) The Personalization of Mediated Political Communication: A Review of Concepts, Operationalizations and Key Findings. Journalism 13 (2): 203–220. DOI: 10.1177/1464884911427802.

Van Dijk T. A. (1993) Principles of Critical Discourse Analysis. Discourse & Society 4 (2): 249–283. DOI: 10.1177/0957926593004002006.

Wodak R. (2020) The Politics of Fear: The Shameless Normalization of Far-Right Discourse. 2nd ed. London: SAGE Publications.

Wojczewski T. (2020) “Enemies of the People”: Populism and the Politics of (In)security. European Journal of International Security 5 (1): 5–24. DOI: 10.1017/eis.2019.23.

Как цитировать: Темир Э., Каракулле И. Дискурсивные изменения в публичной риторике президента Турции Эрдогана (2014–2025): корпусноориентированный количественный анализ // Вестник Московского университета. Серия 10. Журналистика. 2026. № 3. С. 29–69. DOI: 10.55959/msu.vestnik.journ.3.2026.2969

Поступила в редакцию 20.02.2026