Identifying Social and Cultural Trends in British Honours
Despite the horrific news reported on various media on a regular basis, many experts believe that we live in the most peaceful and safe times in the history of mankind. Of course, such sentiments are difficult to prove rigorously, not just because of the fact that there is still much to be uncovered about our past but also, mainly due to the subjective nature of these assertions and the complexity of a potential verification process. Nevertheless, it is reasonable to assume that we do indeed live in very good times; the statistics of life expectancy and infant mortality rates across the world speak for themselves, while material welfare is more ubiquitous than ever by the virtue of economies of scale and industrialisation.
Though this trend is more prominent on the scale of a century or two, the changes in general sentiments and moral values of society are tangible even across a few decades. The most recent series of transformations in values, felt particularly in the west, was a resurgence of populism and protectionism, heralded by the referendum on Britain’s membership of the European Union and the election of the new US president, the cultural effects of which affected the anglosphere most severely. This is while other western nations in Europe experienced a migration crisis, that further fueled populism in those regions.
As the social and political atmosphere is transformed, the moral values of a society and what it honours follow suit. To examine this further and capture the zeitgeist of each period, one approach would be to look at the actions that are applauded in a society, a prominent example of which is the British honours and awards system. Every year, numerous honours are conferred upon the subjects of the British crown that have excelled in their area of expertise, publicly acknowledging their services and applauding them for their actions. Due to the public nature of these awards and their link to the governing establishment at the time of the award, they could serve as a reasonable proxy for measuring that which is valued the most by the British establishment around the time the honours were conferred.
To this aim, a simple analysis, based on word-frequency can be conducted on the descriptions of these awards throughout a number of decades, yielding an indication of what has been honoured in each period. The list of award descriptions can be obtained from wikipedia for each year from 1890 until 2018 for New Year Honours. An issue that presents itself in obtaining data from wikipedia is that not all years are perfectly consistent in their format and the length of the lists vary across the years. This is particularly pronounced during the period from 2003–2006, where the lists are considerably shorter and are presented in a slightly different format, causing a dip in most time series, which will be clearly identifiable in the charts presented later on, in this article.
To begin the frequency analysis on the awards, the new year honours list for every year from 1978 until 2018 was obtained from wikipedia by downloading the webpage for each year and extracting the lists using the XML package in R. With the help of the tm package, word frequency tables were then created for each year, recording the number of times every unique word was used through each year’s list. These tables can then be merged to get a table of frequencies of each word through the years, in the form of a time-series. It should be noted, that to make the results more meaningful, a number of words needed to be removed. The first set of words to be removed were English stopwords such as a, and and by, followed by symbols and punctuation in addition to highly frequent words, specific to the document that do not necessarily contribute to new insights, such as honour, royal, British, queen or services.
After removing the excessive words and characters, term frequency tables are created and normalised to the number of entries in each years honours list, to represent the percentage of the entries that include each term.
An interesting feature of the annual honours that attracts the attention after creating the term frequency tables, is the ubiquity of the name John. For a closer look at the frequency of different names, a few can be plotted against each other as follows.
The issue of name frequency also raises a question regarding the role that gender plays in the honours system. To gain insight into the gender distribution of the recipients, two lists of 1000 popular feminine and masculine English names can be downloaded to find the names that appear in the awards, plotting the aggregate for each gender and normalising to get the gender ratio. It is noteworthy that since people have multiple names (first, middle and last), the sum of the frequencies of unique names does not represent the number of individuals, hence despite the frequency table being normalised before this step, the values almost always added up to more than one before normalising again to display the gender ratio. With this in mind, the graph below may only show that men tend to have more names than women and not necessarily that there are more men receiving the awards; however, since the number of recipients is fairly large and most individuals use roughly the same number of names, it is reasonable to assume that the graph below captures a shift in the gender of the recipients.
This graph clearly indicates that there has been a gradual shift towards honouring more women through the past few decades, with the ratio currently sitting at 0.66 in favour of men, slowly approaching the point of equality, while during the 80’s the ratio hovered at around 0.8. For comparison, the lowest male to female ratio reached in recent decades is 0.64 in 2017.
Moving forward with the analyses, the awards’ descriptions were divided into 5 categories: Sciences and Arts, Military, Government, Social Causes and Business, each identified based on a series of keywords, such as professor, research, arts and drama for Sciences and Arts and charity, voluntary and community for Social Causes. The data across these five categories was then normalised, smoothed using 3-year moving averages and graphed below.
It is evident that the sentiments follow a pattern through the decades with changes in political atmosphere, e.g. a clear rise in the honours awarded for services to social causes, however, further deductions will be left to the discerning reader.
