This study investigated a number of aspects of written sentences and their relationship to overall cognitive ability and mortality. There were 197 subjects from the 1921 Lothian Birth cohort whom had sat a mental ability test at age 11 and a number of other tests relating to intelligence, health and cognitive ability, including the Mini Mental State Examination (MMSE) at a later testing in 1997 or later. Following McCarthy et al’s (2004) experiment, sentence length, polarity, letter case and legibility of the sentence writing component in the MMSE were investigated with regard to their link to overall cognitive ability and morbidity. Estimates of each sentence writer’s health, intelligence, polarity and legibility were gauged by 20 raters, though significantly similar ratings were found only with polarity (r=.58, p<.001) and legibility (r = .52, p<.001). No significant associations between any of the aspects of sentence writing and overall MMSE score were found. Sentence length was the only variable that was significantly associated with survival (HR= .88, p=.04, CI = .78 to .99). These results suggest that, contrary to what much of the previous literature states, the aspects of written language investigated may not be sensitive enough measures to be used effectively in language tests to differentiate between normal cognitive functioning and mild cognitive impairment.