Indiana University - Purdue University Fort Wayne
Opus: Research & Creativity at IPFW
2016 IPFW Student Research and Creative
Endeavor Symposium
IPFW Student Research and Creative Endeavor
Symposium
3-30-2016
Prosody Markers as Phonetic Correlates for
Persuasive Language in ISIS Speech
Kelsie Gillig
Indiana University - Purdue University Fort Wayne
Follow this and additional works at: http://opus.ipfw.edu/stu_symp2016
Part of the Anthropology Commons, and the Modern Languages Commons
Recommended Citation
Gillig, Kelsie, "Prosody Markers as Phonetic Correlates for Persuasive Language in ISIS Speech" (2016). 2016 IPFW Student Research
and Creative Endeavor Symposium. Book 24.
http://opus.ipfw.edu/stu_symp2016/24
This Book is brought to you for free and open access by the IPFW Student Research and Creative Endeavor Symposium at Opus: Research &
Creativity at IPFW. It has been accepted for inclusion in 2016 IPFW Student Research and Creative Endeavor Symposium by an authorized
administrator of Opus: Research & Creativity at IPFW. For more information, please contact [email protected].
Introduction
Kelsie Gillig
(Indiana University- Purdue University at Fort Wayne, Department of Anthropology)
The goal of this study is to determine if prosodic variables
might be utilized to understand the rhetoric of ISIS’s message.
• RQ#1: What makes ISIS so persuasive to those vulnerable to
recruitment?
• RQ#2: Might language, more specifically phonetic features found
in prosodic markers such as pitch, intensity, and duration, play a
role in this rhetorical appeal?
• RQ#3: What can prosodic variables tell us about the persuasive
nature of ISIS speech?
The goal of this study is to determine what makes the Islamic State (ISIS) persuasive from an anthropological linguistic perspective by comparing the
prosody (patterns of rhythm and sound) of a well known ISIS spokesman, Abu Muhammed al-Adnani, and that of a well-known moderate Sunni
cleric, Sheikh Adnan Ibrahim. Specifically, pitch, intensity (loudness) and duration of words are compared between the speakers in order to identify
prosodic features that may appeal to ISIS supporters vs. moderate Sunni Muslims. ISIS is a terrorist organization that diverges from major forms of
Islam (Sunni and Shia) in its jihadist ideologies (among many other variables), while Sheikh Adnan Ibrahim represents a thoroughly distinct ideology
as a follower of the most prominent form of Islam, Sunni Islam that preaches against terrorism. Two samples of oratory from each speaker have
been studied by extracting prosodic markers, using the linguistic software Praat.. The syllabic and lexically segmented data has then been tested
utilizing empirical methods to statistically analyze correlations between prosodic markers of each individual speaker. Results have led to the
identification of divergence in prosodic markers between speakers—specifically in duration and intensity. Al-Adnani’s rate of speech is much lower
than Sheikh Adnan Ibrahim. Essentially, al-Adnani speaks much slower than Adnan Ibrahim. Further, al-Adnani speaks significantly louder than Adnan
Ibrahim according to the derived mean intensity value for each speaker. Finally, results for coefficient of variation of rhythm reveal surprisingly
similar values—both speakers are utilizing a similar rhythmic pattern.
• Acoustic analysis using Praat, a linguistics software tool (Boersma & Weenink, 2005)
• Segmentation of each lexical category
• Once each lexical item was separated into intervals on one tier, a Praat script analyzing
for duration, pitch, and intensity values
• For every pitch and intensity value, a Praat script measured and documented the
minimum and maximum value within each interval, and generated a calculated
mean and standard deviation for every pitch and intensity value in the data
• These results were then utilized in running further statistical analyses of the data
Results
• Normal conversation intensity (loudness) is reported by the American SpeechLanguage-Hearing Association as approximately 60 dB (“Noise,” 2016)
• Prosody is “a general notion in the study of spoken language that covers a range of
suprasegmental phenomena including intonation or pitch, rhythm, duration or tempo, voice
quality, and pausing.” (Berez, 2011)
• Prosody can potentially provide information about a speaker’s attitude (related to emotion) or
Abu Muhammad al-Adnani
Sheikh Adnan Ibrahim
membership in a speech community (e.g. signals of ethnic identity and gender, speech style).
Further, prosody can even tell us something about the emotional aspects of speech
(sometimes related to linguistic meaning within discourse) and the state of health of the
speaker. (Chafe, 1997; Mannell, 2007).
ISIS Abu Muhammad al-Adnani:
• Oratory is “a linguistically self-conscious form of speaking, rich in devices that frame and re- • Born in Syria
frame its use and contexts of interpretation. Many formal devices—including rhythm, pitch, • Official Spokesman of ISIS
pauses, even musical conventions—play a key role in the effectiveness of oratory.” (Kuipers, • Senior leader within ISIS organization
• Subscribes to the mission of ISIS and works to spread its message
1999)
• Utilizes MSA in his speech delivery
The Speakers
• al-Adnani speaks much slower than Adnan Ibrahim
• A proportionate difference of duration mean values (msx1000) between speakers
yields that al-Adnani speaks 1.786037391 (msx1000) slower than Adnan Ibrahim—
almost twice as slow
• Figure 2 points to calculated speech rate (words/minute) to further this claim
• Coefficient of Variation for each speaker was quite large indicating
considerable variation in rhythm throughout the discourse (see Figure
3), but the values are similar, indicating that both speakers vary their
rhythms.
Figure 2: Speech Rate
Moderate Sunni Cleric Sheikh Adnan Ibrahim
• Born in a refugee camp in Palestine
• Received his PhD in Philosophy from the University of Vienna
• Imam of Shura Mosque in Leopoldstadt, Austria
• Speaks out about Islamic matters of the past and present
preaching against terrorism and other issues throughout the world,
utilizing philosophy and logic—sometimes considered controversial
methods of argument
• Utilizes MSA in his speech delivery
The nature of ISIS:
“If one wants to get to know the program of the [Islamic] State, its politics, and its legal
opinions, one ought to consult its leaders, its statements, its public addresses, its own sources”
(Abu Muhammad al-‘Adnani, official spokesman of the Islamic State, May 21, 2012)
“It is a religious group with carefully considered beliefs, among them that it is a key agent of
the coming apocalypse.” (Graeme Wood, Atlantic Monthly, 2015)
•
Religious ideological roots in Jihadism and Salafism within a form of Sunni Islam
Why Does ISIS claim it is a state? (Kuznar, 2015)
•
•
•
•
Claim to be inclusive (IS includes all regardless of race, nationality or tribe)
ISIL Leadership is just: Claim to care of followers, takes counsel, is humble, kind to followers, rewards
performance (in theory)
Claim to effectively provides domestic services
Militarily successful and Capable (at least until recently)
Figure 1: Praat Analyses
• al-Adnani speaks louder than a vacuum cleaner (70dB) while Ibrahim
speaks just a bit louder than conversational
Definitions
Background
Analysis
wa-laqad waʔda-na
PREP
rəbu-na
ʕzə
wə
dʒɪlə
b
nasər
promised-IPL
God-IPL
mighty and
majestic
in victory
‘Indeed, our Lord, the Mighty and Majestic, has promised us with victory.’
Figure 3: Overall Results of Statistical Analyses
Methodology
Qualitative Variables: are used for grouping data for comparison
(e.g. comparing authors, comparing individual speeches, comparing
beginning, middle and end)
• Author: al-Adnani, Sheikh Adnan Ibrahim
• Ideal due to their comparative natures
• Speech: 4 speeches separately analyzed
• Segments Analyzed: Beginning, Middle, End
• From each, 10 minutes of speech was analyzed within different
domains of the discourse
• Lexical segmentation
Quantitative Variables: Means and standard deviations of these variables
will be compared for each of the groups defined by qualitative variables:
• Duration (Syllable Length, ms)
• Pitch (Fundamental Frequency F0)
• Intensity (Amplitude/Loudness, dB)
• Speech rate (Words per Minute)
Conclusion
This study has served two functions:
• Distinctions in the data indicate differences between phonetic factors that may have influence upon the recruitment of moderate and radical
Muslims.
• Variables that might have persuasive value to those vulnerable to radicalization seem to depend on speaking slowly and significantly louder.
Further research:
• This study constitutes a starting point for analyzing how prosodic features of discourse may influence people vulnerable to ISIS recruitment,
and its results are limited due to the small corpus for comparison. Widening the corpus and incorporating different actors would be one
presumable avenue of further substantiating the obtained results of this study.
• Furthermore, an actual test on subjects would be required to explore the actual effectiveness of different prosodic markers on recruitment.
• Incorporating different types of linguistic analyses, such as the addition of semantic and syntactic feature analyses—and thematic coding
thereof, may further shed light on the rhetorical strategies utilized in this discourse genre to recruit vulnerable individuals as well.
References: Berez, A. L. (2011). Prosody as a genre-distinguishing feature in Ahtna: A quantitative approach. Functions of language, 18(2), 210-236; Boersma, P., & Weenink, D. (2005). Praat: doing phonetics by computer (Version 4.3.01) [Computer program]. Retrieved from
http://www.praat.org/; Kuipers, J. (1999). Oratory. Journal of Linguistic Anthropology, 9(1/2), 173-176; Chafe, W. (1997). The Interplay of Syntax and Prosody in the Expression of Thoughts. BLS Annual Meeting of the Berkeley Linguistics Society, 23(1), 389; Kuznar, L. A.
(2015). Daesh’s Image of the State in Their Own Words. In H. Cabayan & S. Canna (Eds.), White Paper on SMA Support to SOCCENT (pp. 27-30). Arlington, VA: Strategic Multilayer Assessment (SMA) Periodic Publication, OSD/ASD (R&E)/RSD/RRTO; Mannell, R. (2007).
Introduction to Prosody: Theories and Models. Retrieved March 01, 2016, from http://clas.mq.edu.au/speech/phonetics/phonology/intonation/prosody.html;"Noise" (2016). American Speech-Language-Hearing Association. Accessed March 17, 2016.
http://www.asha.org/public/hearing/Noise/;Woods, G. (2015, March). What ISIS Really Wants. The Atlantic.