Published online by Cambridge University Press: 17 February 2020
This paper studies the Cascajal Block, a serpentinite artifact with an Olmec text dating to ca. 1000–900 b.c. (Rodríguez Martínez et al. 2006a). It introduces a new line drawing made possible by the detailed documentation carried out by Carrasco and Englehardt, presents a distributional analysis of the inscription in order to assess the sign inventory proposed by Rodríguez Martínez et al. (2006a) and revised by Mora-Marín (2009), and, most significantly, suggests that certain irregularities in the patterns of sign co-occurrence raised by previous authors (Justeson 2012; Macri 2006) are consistent with spelling practices in logosyllabic scripts, showing that the script is linguistically motivated. The work by Carrasco and Englehardt (2015) on the identification and analysis of sign pairings as likely diphrastic kennings is highlighted as crucial to establishing semantic controls needed for any attempt at decipherment; a new, repeated pairing is proposed, consisting of the PELT-PAW signs. Finally, this paper assesses how the various models for the reading formatting and direction of the text could impact the proposed hypotheses.