The Language of Nuclear Security: Language Diversity in Open-Source Internet Searches

Abstract The purpose of this article is to improve understanding as well as analyses of openly available information related to nuclear security. It covers information availability, accuracy, and access. For those interested in open-source intelligence more broadly, this article considers how to identify gaps in a particular information landscape and how these may be explained. Through three case studies (Turkey, India, and Jordan), the impact of native language on information relating to nuclear sources is explored. Specifically, the study highlights the importance of documenting the cultural background of key nuclear security concepts. Many other sectors, such as aviation, can corroborate the fact that effective communication will prevent dangerous incidents. Consequently, this article argues that a more diverse and inclusive approach to language will benefit research, policy, and practice relating to communication in the nuclear field. The article considers new data on the “language of nuclear security,” describing how open-source analysis can be performed to integrate findings from linguistics. The accompanying discussion focuses on the impact of language inclusion and exclusion on work relating to nuclear sources.


INTRODUCTION AND BACKGROUND
Language diversity has various meanings. It can refer to the existence of multiple languages-statistics relating to knowledge, availability, access, and use; and it can also signify the role of language in social and cultural recognition, awareness, and inclusion. A compelling case study covering all these factors can be found in "nuclear language," "our society's largest conscious attempt to communicate across the abyss of deep time." 1 This dramatic quote refers to the importance of making informed decisions about dangerous materials that will persist for tens of thousands of years into the future. This article will discuss new data on language use in the civil nuclear field, describing how open-source analysis can be performed to integrate findings from linguistics. It will highlight the importance of documenting the cultural background of key concepts such as safety and security, 2 which form a classic representation of language confusion. 3 In particular, the article will explore the impact of native language inclusion and exclusion on work relating to nuclear sources. In doing this, the article describes how to identify gaps in a specific information landscape and why these are there. It highlights issues and challenges encountered when developing contemporary intelligencerelated decisions and policies when it comes to considering new open source-related theories and methods relating to security. Particular attention is paid to the language used when creating and consuming this information.
What is "nuclear language" then? It is a highly specialized language with professional standards, phrases, and acronyms. It is used in a different Diversity and inclusion have been receiving increased attention in the security community, attested by recent conference and meeting agendas, scholarship programs and industry initiatives. It is an area that was flagged due to traces of traditional intelligence and army exclusion policies, 7 reports of microaggressions or other forms of bias, 8 and a strong gender imbalance persisting throughout educational programs, application processes, and career progression. 9 While many organizations have now produced high-level policies for diversity and inclusion, further practical support is needed. 10 One avenue to explore in this regard is language. Although diversity means involving people from different backgrounds, language inclusion has not yet played a significant role in either research or practice when it comes to nuclear security. Technical foreign vocabulary is rarely encountered in the nonprofessional context, providing little frame of reference or opportunity for recollection. What is more, nuclear terminology is a matter of professional, adult language development, which means learning is often conscious (voluntary) and slow. Therefore, the nuclear security field would benefit from considering how its unique terminology is accessed and understood. Language knowledge has been linked to equalization of opportunities and economic advancement-removing barriers to development and participation. 11 Experience has shown that awareness and the use of different languages fosters innovation and international interaction. 12 This article will focus on the language of nuclear security that is openly available and accessible; namely, information that can be obtained from open sources on the Internet. While the Internet has brought about new diverse opportunities for sharing knowledge in various languages, many languages are not present online. As a result, there is a "vast linguistic divide"marginalizing communities and individuals with limited or no access to the Internet. 13 This applies particularly to speakers of nondominant languages, who have been noted to struggle to express themselves in culturally meaningful ways. 14 In the nuclear field, this poses a problem to both publishers and analysts of information relating to nuclear security.

Multilingualism in Intelligence Analysis
It is important to note that language use in the nuclear community is functionally stratified. This phenomenon is also known as diglossia: where two or more languages (or varieties of the same language) are used under different conditions within a speech community, often by the same speakers. Choice in expression will depend on activity, prestige, literary heritage, acquisition, standardization, stability, grammar, lexicon, or phonology. 15 Many studies have been done in this area, both on the successful use of multiple languages in one workplace and on cases of misunderstanding. 16 Within this, the nuclear community also deals with biscriptality, or digraphia: the simultaneous use of two or more writing systems for (varieties of) the same language. 17 Script can provide insight into sociocultural shifts and changes in a speech community, such as a country, region, workplace, or professional field. 18 Working in an environment that utilizes multiple scripts requires becoming aware of new language units, adjusting reliance on phonology, adapting eye movement patterns, and becoming used to different orthographic conventions. 19 This can be challenging, as extensive use of various languages and scripts may restrict someone's ability to communicate, diluting a sense of unity. 20 Yet many nuclear projects will function through multiple scripts. A prominent example is the newly established Barakah nuclear power plant in the United Arab Emirates, which is operated from an Arabic-using country through a coalition led by Korea Electric Power Corporation while employing personnel from over two dozen countries. This means that script choice potentially poses a challenge both to those working on site, and to those trying to gain information on the work.
While these considerations are new in the nuclear field, intelligence analysts widely acknowledge the importance of language and writing. 21 The UK Security Service MI5 notes that knowledge of multiple languages is a "vital skill … to keep the country safe." 22 Intelligence agencies generally place a high premium on the language capabilities of officers in certain positions to provide the most complete and accurate signals intelligence picture to policymakers, military commanders, and Intelligence Community (IC) members. To be specific, the U.S. Central Intelligence Agency hires open-source officers to serve as foreign media experts, who review and assess open media sources to deliver high-impact products to the foreign affairs community. 23 However, little research has been done on the presence of multiple languages or the impact of non-Latin scripts on open-source searching, let alone on the impact of writing systems on security. Opensource analysts have remarked that making use of translation tools to convert between different scripts leaves them vulnerable to unintentional error and inaccuracy. 24 Moreover, it is not enough to directly translate and transcribe-nuances differ depending on whether the information was produced by a native or nonnative speaker, and information can be presented in different voices (formal, informal) or make use of homonyms (words with more than one meaning). 25 In summary, by considering diverse and inclusive communication, this research draws on the influence that language has on national security, along with limited information availability about the language of nuclear security, engagement, and response. With OSINT no longer exclusively being used by the IC, it can provide a fresh way to boost transparency with the public and build trust with adversaries. In this research, we use an OSINT approach to investigate language diversity across three different global contexts. This study encourages the IC to consider harnessing the diverse workforce of the civilian nuclear community-with different language proficiency and novel vocabulary-to contribute to new ways of information gathering and investigation in highly sensitive fields. In addition to operationalizing opensource information, our case studies aid the IC in partnering with the civil nuclear community to harvest the immense intelligence value of open-source data safely, effectively, and ethically-thereby evolving analytic capabilities and enhancing international security.

IAEA English and Translation
The nuclear field is not without language resources. The International Atomic Energy Agency (IAEA) maintains a Nuclear Security Glossary 26 and a Nuclear Safety Glossary 27 in English to ensure that terminology and usage are consistent with their publications. However, it is well known by now that the concepts of safety and security do not necessarily translate across languages, countries, or cultures. Sometimes translation is not possible or even damaging; fixing a meaning that may not exist in the first place. An example here is the polyvalence of German Sicherheit, which can be translated into preformed English concepts of either "security" or "safety." In German, the meaning of Sicherheit is only clear from social or political context. 28 For example, nonnative speakers using a dictionary or an online automated translation tool may not gather the full meaning of what is communicated. Such discrepancies "could lead to misunderstandings and frustrations but also differences in the way problems are perceived and responded to." 29 Other examples from different language groups: In Bengali, officially, there are two words to distinguish between security and safety, but people most often use নিরাপত্ তা (nir apatt a) for either. By this, they mean avoiding physical harm. In Bulgarian, cbuehyjcn (sigurnost) is used to describe both car safety and national security, but for personal safety (e.g., in the context of protective equipment) people will use ,eÅjgacyjcn (bezopasnost), whereas for personal security people would use j[haya (okhrana) (in the context of guarding). In Japanese the word for safety is 安 全 (anzen), which connotes measurable ways of assessing risk; but when describing something like food safety, it is often used together with peace of mind 安 心 (anshin), which refers to emotionality and individual perception. 30 安保 (anpo) and セキ ュリティ(sekyuriti, loaned from English) are used in the context of international relations and military.
While the issue of mistranslation-both linguistically and culturally-has been acknowledged by the international nuclear community in theory, reallife case studies and practical solutions remain limited. The 2007 edition of the IAEA Safety Glossary is also available in Arabic, Chinese, French, Russian, and Spanish. Notably, these have not been updated in the past decade, and there are no separate translations of the Security Glossary. What is more, the Arabic, Chinese, and French versions of the Safety Glossary do not include any observations on differences between languages. However, the Russian rendition contains around a dozen specific comments ("Ghbvexaybe gehedjlxbra," "translator's note"). It includes notes on the meaning of words such as "monitoring," "control," "sabotage," and "security." The Spanish edition also covers the translation of "control." In addition, the Spanish text explains that "nuclear" itself is used in many expressions to modify a noun that, from a logical point of view, cannot actually modify in Spanish and "Por consiguiente, pueden ser objeto de malentendido, tergiversaci on o traducci on err onea y quiz as sea necesario aclarar su uso" ["Consequently, these < concepts > can be misunderstood, misrepresented, or mistranslated and their use may need clarification"].

Lessons from Aviation English
Comparisons have already been drawn between the nuclear field and aviation sector. 31 Aviation was one of the first high-risk global industrial fields to adopt English as its lingua franca. 32 While many languages have served as lingua franca throughout history (ranging from Latin to Arabic), in the twentieth century, English became the "global default" for sectors varying from academia to business. 33 Aviation in particular has been subject to a number of high-profile accidents, which has demonstrated the importance of mutual understanding between everyone involved in flight and airport operation-and from its conception, regulations have included communication as a key consideration. In 2003, the International Civil Aviation Society (ICAO) suggested a worldwide minimum English language standard for use in civil aviation to be implemented by 2008. Their report at the time stated that "communications, or the lack thereof, has been shown by many accident investigations to play a significant role in those accidents" and "of 28,000 Aviation Safety Reporting System reports, 70% cite problems related to information transfer." 34 By 2017, the ICAO was able to differentiate between "aviation English," "aviation language," "plain language," "conversational English," "plain English," and "British English," among others. 35 Key problems they identified were "readback-hearback" errors, call sign confusion, reduced situational awareness due to multilingual radiotelephony, and the use of nonstandard phraseology. In addition, they uncovered a lack of first-person narratives, including interviews (most data on language use was submitted via an airline or company mediator); that most available literature and research was based in or on the English language; and that there was a disconnect between training and real interaction.
ICAO recommendations included an increased emphasis on reporting language-related miscommunication issues, stressing that they are as essential as any other issue (mechanical, passengers, etc.), and to increase vigilance of communication, achieving improvement through training. In the case of the English language specifically, ICAO determined that native speakers would equally benefit from training in order to develop the described "aviation English"-a separate professional working language. Whether the nuclear industry wants to adopt one universal language or not, it can certainly learn from the lessons presented in the ICAO report. Almost all of the described challenges exist in some form in the nuclear field. The aviation sector has highlighted that a common understanding of language is essential to safety and security. Perhaps more importantly, they have found ways to improve communication and prevent miscommunication. Particularly emerging nuclear states will benefit from this knowledge, as they can implement language assessment from the outset.

RESEARCH CONTEXT AND METHODS
To investigate language use, choice, and accuracy in the nuclear field, we will present three case studies from different regions and language groups. These were chosen because they all fall under different language groups: Turkic, Indo-Aryan, and Semitic, and use different scripts: Latin, Devanagari, and Arabic. Our analysis will focus on online open-source searches for information about nuclear security. 36 Nuclear materials form a particularly sensitive category of information and, therefore, are a challenge in the opensource arena. We identified what sources of information on nuclear security were publicly available in a given country and language.

The Human Factor in Open-Source Investigation
In our investigation, we wanted to emphasize the human factor in opensource analysis: how someone may search, what is significant, and why they require the information. We considered the accessibility and usefulness of different sources, paying particular attention to information categories, content themes, conceptual accuracy, and publication dates. 37 Since sources of information on nuclear security are scarce (under 100 search results for most countries at the time of writing), the scope of our study was able to cover the complete set of webpages available at the time of the searches (August 2020 and December of 2020, with independent verification in April 2021 and July 2021). 38 In order to guarantee high sensitivity and specificity, we optimized search settings and keywords: we set the search region and language to the target country; we used the "allintext" function to avoid pages linking to unrelated articles or archives; and we noted whether the words "safety" and "security" were used according to the IAEA definitions or otherwise. All categorizations and definitions were verified by a second researcher to avoid bias.
The nuclear industry and policy are part of complex, high-level international processes. Through large organizations that function in English and treaties that are drawn up in English, nuclear decisionmaking is inevitably connected to the English language. English also remains the dominant search language of the World Wide Web. However, this preeminence is a warped reflection of real life: on a national level, users tend to navigate the Internet in their native language, or the language that is most popular in their area. So, when searching for country-or region-specific information, it is essential to consider in which languages the information is required and available. While, as mentioned, this has been acknowledged by the OSINT community, the mechanics of this process remain imprecise. What is available in a given language, and what is missing? Should search methods differ between languages, and how? The countries we selected for case studies were Turkey, India, and Jordan. While Turkey and Jordan are emerging nuclear states, India can be categorized as a mature nuclear state. 39 The goal of these case studies was to establish how someone working in any of these countries may acquire open-source information on nuclear security through an Internet search and to produce advice on how to conduct such a search successfully. We compared which language resulted in a higher quantity or quality of results and examined gaps in information.

Turkey 40
Turkey is known as a "nuclear newcomer." 41 It currently operates a TRIGA MARK II research reactor at Istanbul Technical University, 42 and it is completing its first nuclear power plant in Mersin province. 43 Turkey has a population of around 82 million people, and Turkish is the most widely spoken language-although several important minority languages, such as Kurdish, are also used across the country. English is prominent in schools and certain sectors such as tourism. 44 The translation and meaning of "security" (usually emniyet) and "safety" (usually g€ uvenlik) in Turkish has been discussed previously, and there is some disagreement on interpretation and use. 45 This means it is significant to observe confusion and misuse in published information. For example, webpages that included a flawed understanding of security primarily fell in the "nuclear energy" category (see Figure 1).
In our search, the Nuclear Security Summits were addressed more than any other aspect of nuclear security. 46 These events were clearly of great significance to exposure of the concept in Turkey. In general, over time, there were few results that discussed nuclear security in an exclusively domestic context. News media were responsible for the largest number of results, while academic publications were scarce (even though they certainly do exist). In addition, few results related to the B61 nuclear bombs Turkey stores for the United States at the Incirlik Air Base, or interest in Russia's S-400 antiaircraft missile system-also when the search terms were adapted. This can be explained by Turkey's opaque communication policy when it comes to tactical or nonstrategic nuclear weapons.

India
In contrast, India has developed, demonstrated, and deployed a range of technologies and scientific prowess in the civilian aspects of nuclear sciences. At the time of writing, India has twenty-three operable nuclear reactors, with many more planned and under construction. 47 India has a nuclear energy program, and it is also well known for its nuclear weapons program with a policy of "credible minimum deterrence" and "no first use." 48 Although more than a hundred languages are used across India, Hindi is the most widely spoken language. 49 English is also used as a mother tongue by some, and it is the most spoken second language in the country, as well as a dominant third language. India's strategic nuclear debate, essays, and current affairs are traditionally centered in English-language newspapers. Notably, despite the Hindi word suraksha being the synonymic terminology for both "safety" and "security" in the Hindi language, very few webpage results were seen to misconstrue the words. The majority of the webpage articles and reports used suraksha with a safety undertone when describing the process or situations with inherent risks (i.e., an accident or a malfunction); while the same word had a "security" undertone when used for describing a process or situation involving threats (i.e., intentional, or malicious, from external factors or entities).
Overall, Internet searches yielded notably fewer results than Turkey, and they did not reveal relevant data on regulations, policy, or other important factors relating to nuclear security (see Figure 2). The searches also exposed that the volume of reports from the United States on the governance of India's nuclear security was higher than the reports from the Indian government or academic journalism. As a public open-source user and a civilian, it is difficult to obtain a clear picture from the outside of the security measures in place. Given India's reliance on secrecy, as a vital component of counterstrategy, a social science researcher would need both technical skills and insider access to obtain information that, in principle, should be available to the public. Often, the unavailability of information can be misconstrued for the absence of measures.
The topic of nuclear security, like nuclear weapons, nuclear energy, and military medicine, lacks transparency in India. One motivation for imposing secrecy on the nuclear program involves its restriction to peaceful applications. Secrecy and lack of transparency have resulted in limited information and the inaccessibility of information to the public. Based on the 1962 Atomic Energy Act and Official Secrets Act, the Department of Atomic Energy (DAE) possesses the right to restrict any information related to atomic energy and prohibits the "misuse" of official information regarding facilities relating to the military establishment or energy "works," especially with "foreign agents." The judiciary has also frequently interpreted these laws in favor of secrecy rather than the people's right to know. 50 The results identify bureaucratic inertia from the government and the nuclear establishments, the nature of the topic (nuclear security), and the intertwining concepts of safety and security (in the India region and Hindi language) as parameters impacting the effectiveness of the search results in the public domain.

Jordan
Jordan imports 94% of its energy supply. 51 However, it set up the Jordan Atomic Energy Commission in 2007 to develop a roadmap to a peaceful nuclear energy program. 52 The vision to put Jordan on par with other nuclear countries included building infrastructure and introducing the vocabulary of nuclear energy. This was initiated by establishing Jordan's first nuclear engineering program at the Jordan University of Science and Technology. The university also houses Jordan's Research and Training Reactor, the country's first licensed nuclear reactor, which became operational in 2017. 53 Jordan has a population of around 10 million, and the main language spoken is Arabic. Different versions of Arabic are spoken in different regions, but most people are familiar with Modern Standard Arabic (MSA) and Levantine Arabic. Schools and universities commonly offer and make use of English as well. 54 In Jordan, the majority of English Internet search results were government webpages, followed by career-related content and academia-whereas there were few news media outputs. In Arabic, however, there were more news media results-specifically from Jordanian newspapers targeting local audiences (see Figure 3).
Since the number of search results was low in general, we experimented with different key words, spelling, and grammar (which differ strongly between English and related languages, and Arabic). For example, we noticed a difference when we removed the definite article ' ‫ﺍ‬ ‫ﻝ‬ ' (al-) from our key words. This resulted in slight variations in output: sometimes fewer, but more useful. We dug deeper into the language differences, considering diverging interpretations of key terms. For instance, the word ' ‫ﺃ‬ ‫ﻣ‬ ‫ﺎ‬ ‫ﻥ‬ ' (amaan) is not commonly used when talking about nuclear safety. Rather, experts and practitioners often refer to ' ‫ﺳ‬ ‫ﻼ‬ ‫ﻣ‬ ‫ﺔ‬ ' (salama) or rather '  ‫ﺍ‬  ‫ﻟ‬  ‫ﺴ‬  ‫ﻼ‬  ‫ﻣ‬  ‫ﺔ‬  ‫ﺍ‬  ‫ﻟ‬  ‫ﻨ‬  ‫ﻮ‬  ‫ﻭ‬  ‫ﻳ‬  ‫ﺔ‬ ' (alsalama alnawawia). The word shares a common root with the word for peace ' ‫ﺳ‬ ‫ﻼ‬ ‫ﻡ‬ ' (salaam). In addition, the phrase '  ‫ﺍ‬  ‫ﻟ‬  ‫ﺴ‬  ‫ﻼ‬  ‫ﻣ‬  ‫ﺔ‬  ‫ﻭ‬  ‫ﺍ‬  ‫ﻷ‬  ‫ﻣ‬  ‫ﺎ‬  ‫ﻥ‬  ‫ﺍ‬  ‫ﻟ‬  ‫ﻨ‬  ‫ﻮ‬  ‫ﻭ‬  ‫ﻱ‬ ' (alsalama wal'amaan alnawawi, "safety and nuclear safety") is also commonly used when addressing nuclear safety topics. The terms ' ‫ﺳ‬ ‫ﻼ‬ ‫ﻣ‬ ‫ﺔ‬ ' (salama) and ' ‫ﺃ‬ ‫ﻣ‬ ‫ﺎ‬ ‫ﻥ‬ ' (amaan) are often mentioned together without a specific or intuitive linguistic indication if grouping them adds or alters their individual meaning. These nuances indicate how essential it is to have native knowledge of the search language, as well as local knowledge of the search context.

DISCUSSION
We understand that there are important differences between online and offline language use. Dialects influence how we experience a language; command of a language can differ between speaking, reading, and writing; in writing, people often have a chance to consider what they are communicating

THE LANGUAGE OF NUCLEAR SECURITY
and to edit it (spontaneous versus planned); and we may express things differently depending on the situation (e.g., public versus private).
What we have looked at in this article is a highly specific form of language: one that is primarily used in a professional context by individuals who usually have some knowledge of English. In rich multilingual contexts such as this, distinctive varieties of English have emerged over the years. For example, both "Turklish" and "Arablish" were first recorded in the 1990s. 55 Perhaps most significant here, however, is "Hinglish," a hybrid mixing of Hindi and English within conversations, individual sentences, and even words-which is gaining a lot of popularity within the masses of India. 56 This dynamic was observed in the results of our current analysis, where the usage of the Hindi word suraksha had the appropriate undertone of "safety" or "security" based on the conceptual content and the background knowledge. In conversational speaking, monolingual (Hindi) speakers and bilingual (Hindi/English) speakers are often found to loan words from English to replace common Hindi equivalents. Such borrowings are typical of daily media (and general) usage in India. For example, Prime Minister Modi, in his many Hindi speeches, uses a larger number of English loan words and phrases to refer to concepts that he wishes to emphasize in his political agenda. For instance, when addressing the Nuclear Security Summit Indian initiatives to the Indian public, he used the standard English loanwords or loan translations, such as "pHramebHRk" (framework), "Especialist" (specialist), and "iNTarvyoo" (interview) to adapt more or less to the phonetics of Hindi and accentuate his message. 57 Therefore, when considering the Internet search data presented in this article it is important to remember the "lived experience" of speakers. Mixed English communication styles are quickly growing across the world, particularly in relation to governance and international industry. 58 Next Steps for the Language of Nuclear Security Overall, the three case studies suggest that matching education and training to the specific informational needs of a country's nuclear power program is an important aspect of a successful human resource development strategy. In India, most nuclear-related discourse available on the Internet is centered on English language-based media and newspaper articles, so it makes sense for India to have a large amount of working documentation translated into English. Turkey, on the other hand, has a lower English proficiency in the nuclear industry and also faces a unique challenge of regulator-operator information asymmetry. Turkey's agreement with the Russian Federation, although aimed at minimizing the financial burden of developing a nuclear power program, comes at the cost of foregoing opportunities for domestic workforce development. All aspects of the construction, operation, and regulation of Turkey's Akkuyu nuclear power project will require technical assistance from a Russian vendor. Confusion may arise during technical discussions between Russian and Turkish speaking personnel because it is expected that not all of them would speak a common language. 59 As underscored in our analysis, in order to eliminate misunderstandings and suboptimal performance of nuclear sector professionals and institutions, it is essential to ensure adequate common language skills to be able to communicate with a variety of different stakeholders at different levels of society. Developing a Turkish glossary of frequently used nuclear keywords could help reduce misinterpretations and ensure the accuracy of information exchanged among radiation professionals. 60 In the case of Jordan, the challenge is to make sure that the version of Arabic being used to translate technical documents is equivalent to the language and writing used in a particular region. While the IAEA's documents are mostly translated in MSA, different regions use different dialects, leading to potential misunderstandings that could have negative consequences. 61 Keeping in mind that most personnel working with nuclear materials will be required to use English-for example, those working at the power plants, hospitals, universities, and research labs-bilingual and monolingual Hindi speakers may adapt to using English loan words as a form of communication culture in the professional world of business or academia. However, the same cannot be said for the Turkish and Arabic contexts, where English plays a much less prominent role. The question then is, How do we deal with the gap? The results from each case study country showed a much larger volume of English language than local language outputs, yet significant differences in topics and themes as well as notable mistranslations between languages. Should we encourage organizations such as the IAEA to promote "nuclear English," or do we focus on improving the availability of materials in different languages?

Digital Sociology
Language has not been the only issue of concern in this article-it ties in directly with challenges relating to inclusion. In Turkey, 71% of the population uses the Internet; in Jordan this is 66.8%; and in India 34.4%. Compare this to the United Kingdom, where 94.9% of the population uses the Internet. 62 What is more, during the COVID-19 pandemic the Organisation for Economic Cooperation and Development investigated global student access to computers. They found that Turkey ranked 64th out of 77 countries and Jordan 65th-with no data for India. 63 In general, regular Internet users are also often younger, richer, more educated, and located in more urban areas. This applies not only to those who access the Internet, but also those who produce the information found on the Internet. Furthermore, Freedom House's "Internet Freedom Status" list describes the irrespective of the nature of the topic. On comparing the two regional outcomes on the nonsensitive data searches, one can notice a gap in assessing the extent of digital information in the two countries.
Of course, similar experiments can also be conducted for Turkey and Jordan. What it boils down to, however-and this will come as no surpriseis that there is little information in the public domain on nuclear materials, especially nuclear weapons, in most countries.

CONCLUSIONS
This article connects three unique fields: nuclear security, open-source information, and linguistics. We considered how information is affected by the language in which it is presented, especially in relation to intelligence analysis.
Focusing on the nuances of "nuclear language," we studied the impact that language diversity has on the accessibility of open-source information across the different geographic landscapes. While we used safety and security as an example, there are many words and phrases that could also be considered: the difference between risk and threat, the meaning of adversary, and English terminology that just does not have an equivalent in all other languages (such as "deterrence"). We have looked at other industries, such as aviation and robotics, to consider how they deal with translation and information and whether we can draw any lessons from that. The short answer is yes-there is room for more materials, training, and discussion. We also looked at details such as multilingualism and even the effect of writing and spelling on information availability.
We used case studies to illustrate how someone may struggle to acquire useful information on nuclear security. We examined the significance of gaps in information as well as any miscommunication. The case studies showed that Turkey, which is a "nuclear newcomer" country with Internet that is considered "not free," has produced a much higher quantity and quality of information than India, which has been on the nuclear scene for decades, and has almost seventeen times the population size. Although India indeed has its reasons for this opaqueness, it is a genuine concern that information that can keep people safe and secure is so difficult to come by. The case studies also showed that Jordan, likely because it is not currently developing a power plant, has put out almost no information on nuclear security-which means that now is a good time to get ahead of that. Several countries in the region have plans for nuclear power, which will affect Jordan as well, not in the least in terms of people possibly seeking job opportunities on these new projects.
In summary, openly available information on nuclear security in our case study countries is scarce; less information is available in the local languages, and mistranslation is a source of confusion. What is more, our online searches demonstrated limited usefulness. We worked on a method that could enhance understanding of key aspects in different contexts, such as nuclear security in different national, social, and professional settings-which we hope will be valuable in carrying out intelligence analysis. Ultimately, we want to encourage the nuclear community to look beyond internationally dominant languages, both in acquiring information and in publishing information in the first place, and to contribute to clarifying the language of nuclear security and enhancing global nuclear security practices.