Subscribe to DSC Newsletter

I want to interrupt my "blogging fast" in order to discuss developments to my final programming effort called Elmira. On Elmira, among other things, I hold storylines from fairytales, movies, television episodes, and real-life court cases mostly dealing with abductions, forced confinement, missing persons, sexual predators, stalkers, and serial killers. Consider a movie like "Hostel" directed by Eli Roth. With my apologies, I am about to give some of the storyline here. Given that the movie is about 10 years old, I can't be giving much of a spoiler at this point. At the end of the movie, there is a lot of death. The term "dismemberment" might be more appropriate. Moving backwards in the plot, one should wonder how all of this death occurred. There were some young men in a foreign country doing lots of alcohol and drugs and interacting with prostitutes. For any number of reasons, these guys decided to trust a total stranger in order to meet desirable women. Personally, I think that the progression between meeting desirable women and dying in a foreign slaughter house is perfectly believable. Let's consider the different types of data in the storyline. This delineation reflects how Elmira stores data: 1) at the beginning there is data pertaining to setting; 2) then as the story develops there is a focus on behavioural and event data; and finally 3) as we approach the ending, people die all over the place, leading to data of outcomes. None of this so-called "data" is particularly structured.

There are already some important principles here that I should point out. A good seed in good soil is likely to lead to a good plant. A bad seed in bad soil might yield an unhealthy plant. Er, then a bad seed in a good soil or a good seed in bad soil might result in a so-so plant. My point is this: only certain outcomes can reasonably arise from particular settings. If the "seed" or storyline is planted or rooted in, say, a friendly family-oriented setting, the likelihood of progressing to an outcome involving dismemberment is much reduced. Or maybe that's just my take on the narrative. Elmira has the capability of ontological matching at the level of settings, behaviours, and outcomes. The abstract components for the setting for "Hostel" might be found in other gory films. This is not to say that different "scenery" will necessarily result in different outcomes - or that similar scenery will necessarily lead to the same outcomes. I suppose a bit depends on what is meant by scenery. The story "The Singing Bone" by the Brothers Grimm is set near a forested area and bridge. The real setting is filial jealousy, rivalry, and perhaps contempt, bringing about the murder of one brother in the hands of the other. I remember during my undergraduate years in Environmental Studies, there was always some question of what exactly the "environment" is. It isn't just the natural but also the social environment.

The idea that the environment can influence outcomes is nothing new: a nurturing and supportive environment can help people while an environment of violence and apathy can harm people. Some psychologists - although certainly not all of them - might interpret "social construction" on an internal level, assessing the external by its internal impacts on the individual. People who have disabilities can probably identify tangible barriers in the external environment for example affecting mobility or access to services. A study of data pertaining to setting can therefore contribute to "structural" awareness to deal with "entrenched" or persistent problems. Accepting the environment as an important consideration means acknowledging that we are not totally in control of our lives - that the narrative can change based on the nature of the physical and social structures around us and governing our day-to-day interactions. On Elmira, aspects of the environment can be attributed to specific individuals; or the attributions can be impersonal.

In Hostel, there is what I call a "killing place" or - as per the dialog in the movie itself - a locale where artists have "exhibits." I have already encountered a court case where a "killing place" was found - where the accused tried to lure people inside. It is a pretty good bet that if "killing_place" is included in the setting, some sort of organized murder effort has been devised. Still, killing can only transpire in the killing_place if there are also behaviours of killing or planning for this purpose. Structural data pertaining to behaviours exists in Elmira through the use of a special tagging protocol called the Behavioural Event Reconstruction Linguistic Interface for Narratives (BERLIN). The tagging can be as detailed or general as desired. It takes a fair amount of effort to generate a lot of tagging. But as the term suggests, given that this is a "linguistic interface," I hope to eventually have Elmira extract tagging automatically. I don't see the process becoming automated anytime soon in relation to movies in which a fair amount of data is non-textual. Elmira can locate a plot based on behavioural similarities. Expressed differently, Elmira is designed to find behaviourally similar cases in order to obtain the settings and outcomes. A person can ask, "What sort of circumstances existed before this chainsaw was used to dismember the victim?" Well, in Hostel, the chainsaw scene emerged from the narrative of these dumb guys searching for easy women.

Outcome data is interesting because, more than behaviours and settings, the ontology is asserted. For example, one might assert that a series of behaviours should be regarded as a "bad day." (Being cut up into little pieces is a pretty bad day.) Quite a bit depends on my assertions. Since my master's degree is from a "critical" discipline, perhaps I can assert a great deal. I would even suggest that my specific area of study is particularly useful in cases of predation, oppression, control, confinement, powerlessness, disenfranchisement, exploitation, negligence, incapacity, and abuse. Outcome data is where many objects invoked in the behavioural data are presented in a disassociated state. Setting and behavioural data are "mass associated" with a body of outcomes. I actually find it fairly relaxing designing definitions. Once constructed, I can rapidly determine the general nature of a case without having to reread all of the details; this means that I can safely store many more cases than I can ever hope to remember. I guess my use of definitions might be compared to "bullet points" for a narrative. However, I will show that this is not necessarily a reductive process.

The following definitions have been associated by Elmira with the mass data object file that I have for Hostel's outcomes: Abduct=50 Bloodmoney=100 Class=33 Confine=66 Coverup=12 Death=28 Disability=16 Genocide=20 Gore=84 Labour=100 Rage=20 Regression=25 Savage=50 Scam=15 Vigilante=100. Now, I turn readers' attention to a children's illustrated book called Mr. Maxwell's Mouse: Bloodmoney=100 Class=33 Gore=7 Lie=25 Scam=23 Uprising=16. Both Hostel and Mr. Maxwell's Mouse have definitions called "Bloodmoney" (a business premised on killing) and "Class" (exploitation by the rich and affluent). Although not all people reading this blog have seen Hostel - and perhaps fewer have also read Mr. Maxwell's Mouse - I think most people would agree that a Tarantino film and a children's illustrated story likely have different storylines. Using outcome data, it is possible to determine similarities in narrative that might differ in terms of setting and behaviour. By the way, Mr. Maxwell's Mouse is about a restaurant for cats where mice are bred and served. Mr. Maxwell, a cat, recently got promoted. He would like to have a nice juicy mouse for lunch. It is wonderfully illustrated. And really, thinking in more abstract terms, it is "sort of" "kind of" "a little bit" like Hostel. People are "served" in Hostel, too; this is to satisfy the "hunger" of those that can afford to pay. Elmira noticed that congruence.

I find myself now very interested in stories. One story that I am following - because I find it really interesting - is Donald Trump. This is so much of a story, I'm absolutely certain it will be a movie or Hollywood musical at some point. Sadly, Elmira didn't exist while Rob Ford was mayor of Toronto. I'm not taking sides here. I'm in it for the narrative. Consider the logistical problem of have many stories about the same person on a single or even multiple databases. Is each record a story - or are all of the records a story? When does a narrative start and stop? For example, when Sarah Palin put her support behind Trump, the stock market was struggling. Some would argue that the decline in assets among Americans is part of the Trump story - as the non-attainability of the American dream brings about increased radicalism among voters. Another type of data that Elmira holds is in the form of "spatial intersects." Spatial intersects allow me to stitch together a bunch of records and assert relationships between the records. There doesn't have to be a linear progression for a narrative to be coherent. For example, within a particular narrative there might be a number of different characters. One might have separate records for each character although there could be a confluence of interactions giving rise to timeline intersections. If that sounds complicated, to me it is - if one thinks too big. Thinking small, I just want to make certain that I can reconstruct the broader narrative at a later time.

The final type of data that Elmira can retain in its objects is called a "scene." Conceptually, scene is something that persists over a significant period of time. In practice, I use scenes to hold chronological data where impacts can be expected to last a number of years. For example, one might postulate, a landmark movie such as Hostel or a television series like Dexter might inspire some people that are not well to act out. I could using Elmira determine what court cases appear related to Hostel or Dexter - maybe in terms of setting or behaviours if BERLIN tagging contains enough resolution. The clincher would of course be in finding plot similarities, which I pointed out might not be literal or material but implicit and social. A narrative is a product normative interaction. If we accept that people are not entirely in control of their lives, their actions are to some extent determined by the surrounding environments. In a social environment, there are social norms and conventions, processes and practices, tempo and rhythm. We are like flowing water riding and rippling over tides, waves, and currents. Complete freedom or personal autonomy is an illusion. Consider this picture below that I took from my phone last summer. The ant in the middle of the flowers seems free; but every step it takes is constrained or controlled in some way by the plant surrounding it. In the human environment, more often than not, this "plant" is social in nature. Narrative extends from social construction. It is therefore important to study how much of a storyline is internally conceived (psychological) or externally extended (socio-environmental). There might be abstract congruencies radiating from key events and milestones.

When the Brothers Grimm wrote their fairytales hundreds of years ago, I believe that food shortages were common in Europe. There were so many stories of wolves and kids being eaten. Whether or not these things occurred in real life and inspired the stories, it is important to be aware of influence and confluence. Child labour and factory bondage would soon become common - offering therefore precursors to the emerging labour reality indicative of desperation on one hand and exploitation on the other. It is understandable to have a social scientist try to explain and predict incidents of social deviance and terrorism. What I consider often missing from the process is the use of data-oriented technology - such as Elmira - and a study of narrative as its own particular discipline. I suppose the closest disciplinary background might be "literature." However, my use of literature is not particularly comparable to how traditional scholars have made use of it in the past. I am almost entirely focused on extractable components of social construction. I use literature and its historical context for the development of data objects. I was a literature major once - and rather briefly. During that year, I found a huge focus on the classics. Prominent literature is not necessarily the most useful for understanding social phenomena or for that matter the compilation of data: for example, Mr. Maxwell's Mouse is perhaps more useful for understanding society than The Faerie Queene. I differ, if I had to identify on one thing, because of my technological focus.

Regarding events as products of social construction, data gathering becomes not a hunt for numbers and measurements but rather for embedded constructs in plot. Accepting plot as a source of data, this creates what I would describe as an endless amount from multifarious sources. Even the news becomes data. There was a case recently of a person with a disability being thrown into a dumpster still alive and calling out for help. I thought about this case: bullying, assault or maybe attempted murder, forced confinement. It would seem unusual to me if the perpetrators were non-deviant, but on that day in question they decided to throw a vulnerable person into a dumpster. It also seems interesting that such an act can occur in silence as if it were purely an issue of personal animosity that bubbled over to become an act of aggression. I consider it more likely that the perpetrators expressed contempt for at least this one vulnerable individual, probably had some kind of rationale albeit not necessarily justifiable, and bragged about the incident afterwards. But missing in the narrative is, well, narrative. If the world is interpreted and logged by numbers and incidents, the story would be filtered out. The story is missing from the data system because the data system is not designed to hold or deal with stories. Think of the massive amounts of intellectual capital being lost each day due to this design weakness or vulnerability.

The court cases that I can access online are usually at the "appeal" level. There are many other cases that never get that far. I suspect that the vast majority of cases never even make it to court. There are sexual predators active where the incidents might not even draw the attention of police or family members. The dark world is a silent one - finding expression in our stories and in the "common knowledge." I recall while I was growing up, I was told about a man who was giving hockey tickets away; but the boys telling me about him said that this man expected sex in return. It was like "folklore" among the kids. While there might not be a place for folklore among law enforcement - until an act of deviance becomes quantifiable or material - nonetheless such a story is important. At a later time, this person was finally arrested. There seemed to be some surprise how long he had been molesting children. He operated invisible to the math and metrics. But his story was strong. In fact, I recall some kids were wondering whether "it would be worth it [for somebody but not them of course]" in order to get hockey tickets. (They seemed to be giving it some passing thought.) That's how much we love hockey in Canada, I suppose. The angel Metatron in the series "Supernatural" (not Megatron from Transformers) was said to be God's chief scribe before taking over heaven. He savoured the storylines; and his stories often became predestiny or fate. In J.R.R. Tolkien's Silmarillion, the music of the Valar eventually transformed to real-life existences of the people of Middle-Earth. A terrorist is also a story-teller, probably drawing much more from social constructs than math: e.g. people will go to work in the morning; leave the building to go to lunch or home; watch the evening news; have cameras near the finish line; need time to realize that a jet has been hijacked; require witnesses; a trail of disorganization so people will come realize their own helplessness

So this has been my overview of Elmira. Above is a screenshot of the data-entry screen as it exists today. I think that I can probably spend almost all my free time entering cases. (The court cases themselves are fairly interesting.) One of the benefits of doing my own programming is being able to automate processes that seem repetitive. This is certainly the situation in relation to invoking tags. I don't try to internalize the work. I will never be able to remember all of tags or cases. My description of Elmira should indicate that the software is much more about pattern recognition than compiling numbers. Nonetheless, Elmira can compile numbers. It just so happens that the quantitative value of a qualitative expression is 0 on Elmira. Obviously there is a limit to how useful 0 can be. In relation to following narratives, there is essentially no math. There are lots of algorithmic string comparisons and conditionals. This then is where I have been busily directing my free time. After I have gathered quite a lot of cases, I will make adjustments to the system to allow for more "problem solving." Given the lack of forensic details, I will be focused on the underlying settings and advocating for changes to these settings that give rise to particular behaviours and outcomes.

Views: 341

Tags: assignments, attributes, behavioural, behaviours, berlin, construction, criminology, data, events, evidence, More…fate, features, forensics, gathering, hostel, interface, investigations, linguistic, management, mass, metatron, narration, narratives, objects, pathology, plots, predestiny, reconstruction, scenes, scribe, social, storylines, tolkien, valar

Comment

You need to be a member of Data Science Central to add comments!

Join Data Science Central

© 2019   Data Science Central ®   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service