<p align="justify" class="western">R&eacute;sum&eacute; fran&ccedil;aisLa loi de Zipf s&rsquo;int&eacute;resse aux ph&eacute;nom&egrave;nes de r&eacute;gularit&eacute; dans les diff&eacute;rents domaines de la connaissance. La r&eacute;gularit&eacute; mise en exergue ici est celle de la fr&eacute;quence des mots dans un texte qui s&rsquo;ancre historiquement autour de l&rsquo;ing&eacute;nierie linguistique. Nous pr&eacute;sentons les mod&egrave;les historiques &agrave; travers une formalisation math&eacute;matique commune afin de mieux appr&eacute;hender l&#39;intelligibilit&eacute; des mod&egrave;les historiques propos&eacute;s dans la litt&eacute;rature et de discuter de la controverse entre Mandelbrot et Simon. Nous nous interrogeons sur sa nature et sa r&eacute;silience &agrave; travers une discussion bibliom&eacute;trique et lexicographique. En s&rsquo;appuyant sur la position de Kendall, la conclusion positionnera la loi de Zipf par rapport au SHS.</p> <p align="justify" class="western"><font face="Liberation Serif, serif"><font style="font-size:12pt"><font size="3">Abstract</font></font></font></p> <p align="justify" class="western">Zipf&#39;s law is concerned with the phenomena of regularity in the different domains of knowledge. The regularity pointed out here is that of the frequency of words in a text that is historically anchored around linguistic engineering. We present historical models through a common mathematical formalization in order to better understand the intelligibility of historical models proposed in the literature and to discuss the controversy between Mandelbrot and Simon. We question its nature and resilience through a bibliometric and lexicographic discussion. Building on Kendall&#39;s position, the conclusion will position Zipf&#39;s law in relation to the Humanities and Social Sciences.</p>