For the past several decades, linguistic fieldwork in Libya has been extremely difficult, even at times downright impossible. This has certainly been the case for foreign researchers: not only was it nearly impossible to get research permits for Libya from the 1980s to 2000s, and fieldwork that did occur was heavily monitored and restricted, but there has been so little work on Libya in general, and scholars of Libya in Western institutions, that interested students usually have no place to start or advisors with whom to work. But this also to a great extent true for Libyans as well: Libyans with linguistic training have typically returned to work in universities teaching translation studies or foreign languages and only a few have published research in Libya on Libyan languages. Up until 2011 it was illegal to openly research anything other than Arabic—the regime’s official position was that Amazigh is a dialect of Arabic, and numerous researchers (not to mention activists) were thrown in jail for trying to write, teach, or research Amazigh in Libya. And now, although the activism and dedication of numerous Libyans has led to the increased visibility of the Amazigh and Tebu languages in Libya, actual fieldwork and research remains difficult for everyone due to the current political and military struggles.
I’ve always assumed that fieldwork during the colonial era and during the kingdom was, in contrast, much easier. Foreign researchers could simply have taken advantage of colonial power structures to go where they wanted, and indeed many did. Or after independence they were given permits to do so. And this is largely the case for research on Libya up until the early 1970s in a variety of fields—anthropology, linguistics, history, urban studies and so on. But, on examining a bunch of older linguistic works more closely, I was surprised to find that many of them were not actually done in the place the language was actually spoken at all—some of them not even in Libya. Of course, these studies were still carried out within colonial power structures. But, it’s likely that French scholars, for example, weren’t as easily able to travel to then-Ottoman Libya as they were able to travel within French colonial domains, and therefore took advantage of what opportunities they had to produce knowledge on the region. I’ve gathered some of these sources together under the rubric “non-site fieldwork”, the opposite of “on-site fieldwork”.
§1 Nefusi in Mzab and Ghadamsi in Oued Souf (Algeria)
One of the very first linguists to publish on a Libyan language was the French Orientalist Gustave Adolphe de Calassanti-Motylinski (1854-1907), who spent his entire life in colonial Algeria working as a military translator and then professor of Arabic. Having already carried out research on the language and customs of Mzab in eastern Algeria, he became interested in the Amazigh varieties of western Libya. In Algiers he had been a student of the main Berberologist of the time, René Basset, who put him in touch with a scholar from Yefren named Brahim bin Sliman Chemmakhi (الشمّاخي) who was resident in Mzab. Calassanti-Motylinski encouraged Chemmakhi to write something about the Nafusa Mountains in Nafusi in Arabic script. This written chronicle was published in 1885.  Calassanti-Motylinski’s linguistic analysis of Chemmakhi’s text, together with additional elicited grammatical information, was only published much later, in 1898.  In the preface to that work he admits that his information should be checked in the Nafusa region itself against multiple speakers.
Calassanti-Motylinski then turned his attention to Ghadames. As he writes in a trip report entitled “Notes on a mission to Oued Souf to study the Berber dialect of Ghadames” published in 1903,  he was made aware that a Ghadamsi merchant named Muhammad ‘Uthman al-Ghadamsi would spend several months a year doing business in Oued Souf. Drawing on an earlier relationship with officials in nearby Guemar, Calassanti-Motylinski sent the head of the Tijani zawiya there a wordlist in Arabic to have Muhammad ‘Uthman translate into Ghadamsi. Several months later he received back texts and transcriptions whose quality he was dramatically impressed by. Realizing that he would have the chance to meet with a Ghadamsi speaker who understood the goals of a linguist, he obtained permission to travel in Oued Souf and set off from Constantine in March 1903. Once arrived, the head of the local Bureau arabe, a certain Captain Bussy, put at his disposition several locals who lived in Ghadames and knew the language (the aforementioned Muhammad ‘Uthman being unavailable apparently). According to Calassanti-Motylinski’s report, his best information on the Ghadamsi language actually came from a Touareg servant in the service of a merchant named Muhammad al-Dalw. Though he doesn’t give her name, he notes that she dictated a number of texts in Ghadamsi and was able to explain many unclear or previously unknown words. He spent his last days in Oued Souf gathering popular songs in the local Arabic dialect (never published, I think), and returned to Constantine in mid-April 1903. His linguistic description of the Ghadamsi language based on the materials he collected in Oued Souf, Le dialecte berbère de R’edamès, was then published in 1904 by the École des lettres d’Alger. 
§2 Touareg of Ghat in Tataouine (Tunisia)
The most recent detailed description of Libyan Touareg, and indeed any language of Ghat, is over 100 years old. But it also has the distinction of being quite possibly the first modern linguistic study of a language of Libya to be written by a native North African. Mohamed Nehlil, who usually published simply as “Nehlil”, was a Kabyle Algerian who served in various capacities in the colonial administrations of Algeria and Morocco. He was an officer in the Bureau des Affaires Indigènes in Algeria, taught Kabyle in Algiers where he came to know and study with René Basset, and, after its founding in 1912, taught in the École Supérieure de Langue Arabe et de Dialectes Berbères in Rabat. In 1906, he spent some time working at the Bureau des Affaires Indigènes in Tataouine, southern Tunisia, where he met a Touareg merchant from Ghat named ‘Ali bin Ahmad bin Mhammad Goundi who had been living in Tataouine for several years. On the encouragement of Basset, Nehlil gathered some stories and other texts from Goundi and published them, together with a grammatical description and French-Touareg vocabulary, in 1909 as Étude sur le dialecte de Ghat (with a dedication to Basset).  Nehlil went on to publish a few other studies on North African topics, though no others relating to Libya.
§3 Saharan Languages in Tripoli (Libya)
With the Italian occupation of Libya in 1911, research on all aspects of Libya increased dramatically. Italian (and other European) scholars could simply go where they pleased and study whatever they wanted to. The creation of colonial offices and military bases all around the country meant that, like in French Algeria, a network existed for researchers to tap into. Rather than being limited to the odd European traveler, the Sahara could be accessed by European researchers who could easily obtain accommodations and supplies from other Europeans, and, if necessary, compel local populations to take part in their studies. Hence, “on site” research on previously unknown or poorly-known places such as the Nafusa Mountains, Murzug, or Awjila flourished, as did work on the previously poorly studied Arabic dialects of Benghazi and Tripoli. Even so, some researchers were not able, or did not desire, to carry out “on site” fieldwork, despite the colonial power structures that would have enabled it.
Tomasso Sarnelli (1890–1972) was an Italian opthalmologist-turned-linguist. In the 1920s, Sarnelli made use of his time working in Tripoli to pursue linguistic interests, having been encouraged by Francesco Beguinot (1879-1953), the main scholar of Libyan Berber at that time and professor at the University of Naples (who had criticized the Amazigh people for their “inability to progress beyond the basic levels of civil life”). In Tripoli Sarnelli had the chance to meet an older man from Sokna, Shaykh Hassuna bin Muhammad al-Dakshi, who still spoke the Sokni language. Shaykh Hassuna told Sarnelli that only 4 or 5 people could still speak Sokni, with about another 40-50 having passive understanding. He wrote out five folk tales in Arabic letters for Sarnelli in the fall of 1923, which Sarnelli transcribed into Latin script, translated into Italian, and added a Sokni-Italian glossary to. His article “Il dialetto berbero di Sokna” was published in 1924 with the blessing of Beguinot in the colonial journal Africa Italiana, and is still the only firsthand linguistic study of that language.  Sarnelli didn’t publish further work on languages of Libya.
A few decades later, Umberto Paradisi (1925-1965) began to publish work on Libyan languages. He spent time in Awjila in eastern Libya, where he published what are still the only fieldwork-based accounts of the now-endangered language of that oasis. He also published on prehistoric rock art of the Libyan desert. As he lived for a time in Tripoli, he decided to search for any speakers of hitherto undocumented Berber languages. In 1955 he was able to meet a person in the Sug al-Jum‘a neighborhood whom he refers to as “one of the šewâšena” of the central Libyan oasis of el-Fogaha (about 200km south of Sokna), presumably meaning a Black person (perhaps previously enslaved) who had lived there. This person apparently knew a few words of the local language, and could confirm to Paradisi that it was still spoken. Then, in summer 1960 and early 1961, Paradisi was able to meet 2 older men who were apparently 2 of the last 3 speakers of the language: Muhammad bin ‘Uthman Zaydani of the Awlad Ali and Sharif Yahya of the Awlad Bilgasim. Based on stories and wordlists they provided, Paradisi published two studies in 1961 and 1963, constituting here as well the only fieldwork-based accounts of the (now probably extinct) language of el-Fogaha. 
To give you an idea of just how far away the places where the research was conducted are from the places where the languages are actually spoken, here’s a map. You’ll notice that the shortest distance is 270 miles—a great distance for Saharan travel in the late 1800s and early 1900s!
Since several of these studies remain the only ones of their kind, our information on these languages, though valuable in certain ways, is not only outdated now but was from the very beginning incomplete and out-of-context. (Besides, for example in the case of Sarnelli, being framed in quite problematic and condescending ways.) Updated work in collaboration with speakers is likely to yield a radically more complex picture than what we currently have, and that is a big part of why new research, particularly by local researchers who have better access to their speech communities, is absolutely essential.
I should note, of course, that “non-site” fieldwork isn’t by default a bad thing. Sometimes a community of speakers is indeed dispersed around the world, because of persecution or migration. Sometimes work with speakers outside of their place of origin is a prelude to on-site fieldwork and an important part of making connections with the community. Or, it can follow on-site fieldwork as part of collaboration with local researchers. Those are all good things.
But when a language is still widely spoken in its homeland, research with a single speaker far away from that place is not likely to give the best, most nuanced picture of that language, and moreover, is much less urgent, since that language isn’t likely to die out or be replaced. Strangely enough, the impulse to do “non-site” fieldwork seems to be growing among Western scholars—but for Arabic dialects of northern Africa rather than for the minority or endangered languages which are in need of documentation.
 de Calassanti-Motylinski, Gustave Adolphe. 1885. Le Djebel Nefousa (Ir’asra d’ibriden di drar n infousen): Relation en Temazir’t de Djebel Nefousa composeée par Brahim ou Slimane Chemmakhi, taleb d’Ifren. Algiers: Jourdan.
 de Calassanti-Motylinski, Gustave Adolphe. 1898. Le Djebel Nefousa. Transcription, traduction française et notes avec une étude grammaticale. Paris: Ernest Leroux.
 de Calassanti-Motylinski, Gustave Adolphe. 1904. Le Dialecte berbère de R’edamès. Paris: Publications d’Ecole des Lettres d’Alger.
 Sarnelli, Tommaso. 1924. Il dialetto berbero di Sokna: materiali lessicali, testi manoscritti in caratteri arabi, con trascrizione e traduzione. Supplemento all’Africa Italiana. 1–43.
 Paradisi, Umberto. 1961. El-Fógăha, Oasi Berberofona del Fezzân. Rivista degli Studi Orientali 36. 293–302.
Paradisi, Umberto. 1963. Il Linguaggio Berbero di El-Fógăha (Fezzân). Istituto Orientale di Napoli 13. 93–126.
** Honorable mention goes to Hans Stumme (1864-1936), whose Tripolitanisch-tunesische Beduinenlieder (1894) is actually based on songs recited to Stumme in Tunis in 1888 by one Bilgāsem originally from the city of Matmāta in southern Tunisia.