ASE: The Argument Search Engine

Your search engine provides you with relevant results. It lists everything out by popularity and usefulness and relevance and web history. It provides suggestions. It recommends restaurants and local car dealers. It links news and blogs and Twitter feeds.

My search engine refutes yours. Introducing…

Argument Search EngineThe Argument Search Engine (ASE) is a search engine that provides you not with relevant arguments, but with relevant counter-arguments. It is an engine that tries to disprove your search phrase, criticise it, and propose alternate theories. The ASE challenges your point of view and promotes freedom of thought.


The idea was born from a comment on a TED Talk several years ago. I was conducting research for a debate on the benefits vs. dangers of the Internet, and had found Evgeny Morozov’s talk on “How the Net aids dictatorships,” when I stumbled across this comment by Nic Hautamaki :

“In the beginning, when the Internet was just starting to take off, everyone thought that it would be this huge mind-opener and enlightener. But in actual fact, if anything, the opposite is true. I would say that for many people, the Internet has actually become a mind-closer. What seperates the internet from radio, television, and print is the level of control that users have over what content they view, and the total amount of content available. Since both categories approach infinite with the internet, the result is that people can always find plenty of evidence and support for whatever ignorant preconceived notions they happened to have. People almost never do a Google search on something they have no opinion on; they do a Google search in order to find justification for whatever belief they already had. The end result is that instead of someone with some vaguely incorrect idea, you have someone that’s 100% certain of their incorrect idea, and become active advocates for it.”

The concept that Mr. Hautamaki is presenting is not new; it is known as confirmation bias. As Mr. Hautamaki explains, it leads people to favor information that coincides with their preconceptions. People tend to enjoy contributing to communities that share their worldviews, and thus visit them more often. By doing this, they limit their own exposure to radical, new ideas.

The goal for the (hypothetical) Argument Search Engine is to counteract such affirmation of falsities. The ASE would promote critical thinking, fuel discussion, encourage discovery, and hinder the spread of misinformation and misconceptions. You would be able to browse your regular forums, share ideas, and then plug them into the ASE to discover alternate theories. You would be able to create an outline for an essay from one point of view, then type in your main arguments into the ASE and receive the most relevant and valid counterarguments, allowing you to flesh out your paper. You would be able to search for arguments in preparation for a debate. And, if you were bored, you would be able to challenge yourself by simply typing in theories into the ASE, and reading up on the criticism.

Perhaps even more importantly, Internet vigilantes would be able to easily search out misconceptions, and help correct them. Typing in something obvious to you into the ASE could return results with people to whom the idea is perhaps not so obvious. By replying or commenting, you could not only help solve problems, but contribute to the education of the Internet population as a whole.

Now, the most pressing concern would be to figure out how such a search engine would function. In practice, it may take quite a bit of work to implement (though, current high-profile search engines with fleshed-out algorithms could provide it as an option). In theory, however, the method is not too complicated.


There are two possible approaches to searching for arguments to a search term. They both involve text analysis.

The first analyses the search input. It searches for basic verbs (is, has, does), modifiers (nice, good, few, strong) and conjunctions (and, or, nor) and, quite simply, reverses them. So “the sun is green” would become “the sun isn’t green,” “the strong global economy” would become “the weak global economy,” and “rich and happy” – “rich not and happy.”

The second approach analyses the search output. Not unlike OpinionCloud, the ASE algorithm would look for negation and argument amongst the search results. While it is still difficult for computers to analyse such a complex construct as language, using this in conjunction with the first method should provide the users with good results.

Of course, an argument is useless if it is some biased rant on a secluded, anonymous forum. Thus, the reputation of the source would still be a priority in the ASE search algorithm. The validity could be ranked based on the type of content – websites publishing scientific journals and articles would be favoured above forum posts, comments, and video responses. However, the user would have the option to choose for the type of content they wish to see. If they were interested in user discussion, they could check to specifically search only forums and comment threads. If they wanted speculative, but (possibly) more reputable information, they could choose to search only blog posts and personal websites. Finally, they could choose to search for the most reliable content – that from scientific journals and reputable, often-linked to, popular websites.

Depending on the reason why the user was conducting the argument search, they could also specify to sort the results by tone. A ‘mild’ option would provide results that include few words of harsh criticism, and are seemingly unbiased. Selecting ‘critical’ would provide results that comprise of articles and reviews that are still unbiased, but both argue and support the search term. Finally, a ‘fanatic’ option would let you search for biased and hostile content.

Closing Remarks

In the increasingly connected world that we currently live in, we have been exposed to amounts of information that we cannot possibly manage to process. That is why search engines such as Google and bing have become the standard way for navigating around the web – they empower us to find exactly the information we want and need.

However, as the old saying goes, with power comes responsibility. It is our responsibility to weed out the valid and important information while discarding the useless and false. Good judgement and Internet experience usually suffice, but we are all but human, and we make mistakes. Very easily and completely unknowingly, we can slip into the trap of confirmation bias.

This is the ultimate goal of the Argument Search Engine: to dispel misinformation and fuel critical thinking. It is but another tool, one that can return both useful and useless results. However, it is another venue of processing the vast pools of information that we browse through, and it is a venue that has not really been explored to any great extent. If such an engine were to be implemented, it would surely prove to be useful to a significant subset of the Internet population. It would help reduce the amount of falsities and misconceptions floating throughout the network pipes. And maybe – just maybe – it would help direct the Internet into becoming something closer to the mind-opener that it was meant to be from the very start.


About JustasM

Justas is one of the breed of humans that wake up snotty early in the morning, finish whatever scribbling needs to be done, rush out of the front door with a half-formed sandwich seeping radioactive goo clenched in their fingers, run a distance the equivalent of several obstacle courses, try sucking up knowledge like a Slurpee all morning and afternoon, then check off a few responsibilities from a tome, read a textbook or five, robotically string together several millennia of words on a contraption understanding only 0's and 1's, perhaps consume some edibles if they are within arms reach, then finally collapse on the nearest unoccupied flat piece of furniture. He also enjoys life.
This entry was posted in Concepts and tagged , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s