Investigating Google’s Search Engine – Ethics, Algorithms, and the Machines Built to Read Us, reviewed

Latest book review by Kirstie Affleck  who is a Sheep farmer, based in West Cork, Ireland.

Investigating Google’s Search Engine – Ethics, Algorithms, and the Machines Built to Read Us – by Rosie Graham, published by Bloomsbury

This book leads the reader through historical references of memory learning and recall, including reference to Greek philosopher Plato and the Socratic dialogue between Meno and Socrates about what virtue is and whether enquirers can ask questions about topics they are ignorant of, and its relevance to search engine searches.  Studies have shown that we remember things differently if we believe the information will be readily available again, or not.

Investigating Google’s Search Engine – Ethics, Algorithms, and the Machines Built to Read Us, reviewed

The suggestion is made that a search engine can be used as an external memory bank, and that we learn how to find and extract information from the search engine, rather than storing the information in our own brains. Graham delves into the history of memory and machines going back to the Ars Memoria, the basis of modern mnemotechniques, and Ramon Llull who wanted to create a system to convert non-Christians to Christians. In a similar way to how Llull’s Art could be described as a ‘black box’ system, Graham states that Google has the same mysticism, in that if the mathematical algorithms are tweaked the search results would likely be different. 

Graham explains how using a search engine to look up all sorts of things has become such a fundamental part of daily life, and how we search often isn’t addressed fully in academic research of searches carried out. The author explains how search engine enquiries are limited through a lack of information and the user’s personal histories, and how researchers are limited by a lack of access to data given to them by Google. 

Rosie Graham teaches us how searches cannot be replicated as Google Algorithms change on the back of each search. Graham lays out their own experiment very clearly showing how they went about creating a scientific experiment that cannot be replicated, given the variables of a non static Search Engine. Graham explains how the  compulsory Autocomplete drop down suggestion box works and how users are led to believe that the results given in the drop down box are thought to be the most searched for phrases, and how it can have an impact on the user’s thoughts and attitudes, even if the suggestions were not followed.

Bias in Google search results are shown, in particular in this book dealing with sexist, racist, and LGBTQ prejudice. Graham’s research shows how results vary through location and language, and also by year as Google algorithms change, and how some search results become removed once their detriment is pointed out to Google. Graham covers how Google is increasingly reliant on machine learning neural networks and using AI to scan books into the system.

The ethical questions then become whose voice is reaching users when making an enquiry? Which language were the books written in, from which country, and which gender, ethnicity, and position in society did the author come from? And is it possible that the AI is combining information to create its own search suggestions?

Graham explains clearly how Google is monetised using Google Adsense and Google Adwords, and how since Google gets paid by a business customer when an ad is clicked on, rather than an unsponsored search result, both Google and the business customer have a strong incentive for the ad to show up with most relevancy in a users search list. 

Adwords allowed for bias allowing advertisers to select who the ad could be shown or not shown to using various criteria including gender and ethnic groups. I was dismayed to learn that it is believed 95% of languages have already lost capacity to ascend digitally since there is no incentive for advertisers to pay Google for specific words and phrases, leading to an acceleration of online language death. 

I found it especially interesting how a small town in Macedonia was able to capitalize on Google’s system of monetizing advertising during the 2016 US elections and make well above normal Macedonian salaries by circulating Fake News. Whilst misleading, this practice is not considered criminal. 

This is a well written and interesting book for anyone with an interest in how Google Search Engine works, the psychology of memory and mind control, or even on how to monetize a website, although I’m not sure that last point would be Graham’s intention! Having viewed Google Search Engine as a neutral tool to be used freely and nothing more, this book was a real eye opener for me.

Kirstie Affleck is a Sheep farmer, based in West Cork, Ireland. She is passionate about wool crafts especially knitting and spinning, and sells handcrafted products locally. BSc Biological and Medicinal Chemistry, Exeter, UK.  Instagram: @capeclearsheep Twitter: @CapeClearSheep. See her website here.

See more book reviews here.