And now for something completely different…

I will discuss a recent court case in the Netherlands, in which forensic phonetic expertise was called upon to help settle a dispute over trade name infringement. In 2014, Dutch brewer Grolsch launched a beer called Kornuit /kɔrˈnœyt/. Recently, supermarket chain Lidle released a beer under the name Kordaat /kɔrˈda:t/. I was asked by Grolsch to shed light on the phonetic similarity between the brand names. Using the Levenshtein distance metric (Levenshtein 1966, Heeringa 2004), the phonetic difference between the names is 29 percent. To show that the similarity between the brand names was very likely to be intentional rather than accidental (as Lidle would have it), I established the statistical distribution of the similarity of Dutch word pairs. I selected the 3000 most frequent mono-morphemic content words from Baayen et al. (1995) and computed the Levenshtein distance for all 4,498,500 non-identical word pairs (using Gabmap software, Leinonen et al. 2016). Distances ≤ 29% occur in .5 percent of the word pairs, which arguably shows that the name Kordaat was not accidentally chosen by Lidl. In my talk I will explain the Levenshtein metric and motivate the decisions made to obtain the distribution of distances between Dutch word pairs.

References
Baayen, R. H., Piepenbrock, R. & Gulikers, L. (1995). CELEX2 LDC96L14. Web Download. Philadelphia: Linguistic Data Consortium.
Heeringa, W. J. (2004). Measuring dialect pronunciation differences using Levenshtein distance. Doctoral dissertation, University of Groningen.
Leinonen, T., Çöltekin, Ç. & Nerbonne, J. (2016). Using Gabmap. Lingua, 178, 71-83.
Levenshtein, V. I. (1966). Binary codes capable of correcting deletions, insertions, and reversals. Soviet Physics Doklady 10(8), 707-710.