In a world where communication and sharing of thoughts across the globe is constant and near instantaneous, there is much insight towards human culture that can be extracted from what people say and write (Michel et al, 2010). Through analysis of the frequency of words contained in over five million books spanning hundreds of years, a culturomics study carried out by researchers from Harvard allowed the inspection of cultural trends in human history and gave us the ability to use those trends as legitimate data for further study in disciplines such as lexicography and linguistics (Michel et al, 2010). With the use of optical character recognition technology sponsored by Google, millions of books with publishing dates that span hundreds of years
…show more content…
It is claimed that culturomics studies neglect many factors that make a language a language, such as syntactic patterns and the way words sound; and that a language is not simply the words that make it up. For that reason, some linguists assert that the detachment from detail that is inherent in culturomics studies makes any findings in linguistics unreliable (Zimmer, 2013). Other linguists criticize the dataset itself, claiming that the optical character recognition scanner makes many mistakes; and the different ways of spelling a word, capitalization and grammatical inflection makes the data ‘noisy’. Lastly, some of the linguists in Zimmer’s article say that many of the findings are nothing new, and have been known concepts in the linguistic discipline for a long …show more content…
For instance, individual words lack context that can provide meaningful reference as to why a word is frequent or not. To elaborate on that, searching for the words ‘Google’ and ‘Apple’ to see which has a bigger cultural impact became difficult because of the many health and nutrition books about fruits, specifically apples that dated back as far as 1576 in the Ngram viewer. On the other hand, the trend for ‘Google’ accurately showed that the first peak began near 1997, which is when the company was founded. Next, the 2-gram ‘Apple Computer’ was compared with ‘Google’ in order to work around the issue. However, it is not very often that the company is referred to as ‘Apple Computer’ compared to simply ‘Apple’. Therefore, the resulting trend line did not seem to portray an accurate representation of the cultural impact by the company Apple. That is not to say that there is no value to be found in the Ngram viewer, as many words do not need that extra context. For example, an interesting query to give the Ngram viewer is ‘car’ and ‘carriage’. By plotting these trend lines, the history of transport and the overtaking of horse-drawn carriages by motor vehicles becomes very clear on a time line point of view. It is apparent that in the early 1800’s, ‘carriage’ was widely accepted as the one way for
It is claimed that culturomics studies neglect many factors that make a language a language, such as syntactic patterns and the way words sound; and that a language is not simply the words that make it up. For that reason, some linguists assert that the detachment from detail that is inherent in culturomics studies makes any findings in linguistics unreliable (Zimmer, 2013). Other linguists criticize the dataset itself, claiming that the optical character recognition scanner makes many mistakes; and the different ways of spelling a word, capitalization and grammatical inflection makes the data ‘noisy’. Lastly, some of the linguists in Zimmer’s article say that many of the findings are nothing new, and have been known concepts in the linguistic discipline for a long …show more content…
For instance, individual words lack context that can provide meaningful reference as to why a word is frequent or not. To elaborate on that, searching for the words ‘Google’ and ‘Apple’ to see which has a bigger cultural impact became difficult because of the many health and nutrition books about fruits, specifically apples that dated back as far as 1576 in the Ngram viewer. On the other hand, the trend for ‘Google’ accurately showed that the first peak began near 1997, which is when the company was founded. Next, the 2-gram ‘Apple Computer’ was compared with ‘Google’ in order to work around the issue. However, it is not very often that the company is referred to as ‘Apple Computer’ compared to simply ‘Apple’. Therefore, the resulting trend line did not seem to portray an accurate representation of the cultural impact by the company Apple. That is not to say that there is no value to be found in the Ngram viewer, as many words do not need that extra context. For example, an interesting query to give the Ngram viewer is ‘car’ and ‘carriage’. By plotting these trend lines, the history of transport and the overtaking of horse-drawn carriages by motor vehicles becomes very clear on a time line point of view. It is apparent that in the early 1800’s, ‘carriage’ was widely accepted as the one way for