The Fallacy of Instagram & Alpine Accidents

When Miha Sumi fell to his death on Mount Hood in February of 2018, most people were tuned in to the dramatic rescue attempt live streamed by KOIN-TV helicopter. I, however, was glued to the comments by the outdoor community. Less than a year prior, John Jenkins died on the same route on Hood and I noticed something peculiar about the way people reacted to the accident. There was a lot of grief, a lot of reckoning over the fact that our hobbies put us at unnecessary risk, but also a lot of fascinating mental gymnastics by people who tried to avoid that reality. They create a narrative that accidents happen to the stupid, the unprepared, and the inexperienced, and would therefore never happen to someone like them who’s smart, geared up, and well-practiced. From that latter group, there always seems to be a monumental asshat who goes online and says something to the effect of “This is what happens when people get push past their limits for an Instagram post.” (Never mind that this comment is generally made before they’ve disclosed the victim’s identity, and both their outdoor credentials and their social media usage are unknown. And in Miha Sumi & John Jenkins’s cases, both were experienced mountaineers with prior Hood summits under their belts).

It’s a (fucking bullshit) narrative that I noticed circulating in print media as well, so I got curious about where it came from. Yakima Herald dropped an article two days after a fatality on Mount Adams in 2018. High Country News published a similar feature two weeks earlier, trying to connect a fatality in the Maroon Bells to YouTube and Snapchat. They both cite their theories from a guy named Jerry Isaak, a professor at SUNY-Plattsburg. So the concept has legit, academic research to support it? Hardly.

Isaak introduced the idea that social media was a dangerous presence in outdoor pursuits in 2013 with an article in Avalanche Review. Article aside, it sounds plausible. I’ve seen Insta stories of friends riding conditions I don’t, experiencing FOMO knowing they made it out alive, and only learning offline after the fact that they heard woomphing while they were out.

But that’s not what Isaak’s reasoning is. His stance is that social media’s effect duplicates the social proof heuristic in FACETS. For those of you who aren’t snow dorks, FACETS comes from a data-intensive study by Ian McCammon, who splits his career between being a Ph.D. holder in mechanical engineering and working in outdoor leadership. His research combed through avalanche accidents and codifying 7 major heuristic traps (abbreviated to FACETS) that tended skew people’s risk assessment. Social Proof is tendency to assume a slope is safer if we see other parties skiing the same terrain. That if the slopes were truly risky, there wouldn’t be so many people traveling through them.

So first off, avalanche accidents and the climbing/mountaineering accidents described in the Yakima Herald and High Country News are not one and the same. FACETS just isn’t applicable. The F, for example, stands for familiarity. McCammon found that if we’ve skied and survived a route many times, we tend to underestimate the risk and the slopes we ski most often can become more dangerous. For hiking and climbing, familiarity is generally a positive. You know how long things will take, how many snacks to bring, what pieces of pro you should have along and the spacing between your rap stations. FACETS is so unrelated to rock climbing and scrambling (the activity du jour at the Maroon Bells incident) that Google thinks you’ve made a spelling error when you pop the two in the search bar.

But even within the scope of avalanche accidents, Isaak’s logic still doesn’t check out. He argues that McCammon’s research, when published in 2003, social proof was the result of the presence of nearby groups. His stance is that today’s millennials are so social media obsessed that we can no longer differentiate between physical presence and virtual presence. And he goes on to quote a WIRED article that millennials “make no distinction between the real and the virtual. Actions that begin in one realm play out in the other. They are interwoven.” As a millennial, I call bullshit on this. Mainly because I’m pretty sure I know the difference between what’s on my phone and what’s in my surroundings. But also, the concept of social proof requires that the other parties are engaged in similar activities. The people watching my Instagram stories are by and large not traveling in avalanche terrain. They’re riding a bus or waiting in line at Target or in bed scrolling before they hit the hay. None of which are even risky situations.

So I also tracked down the WIRED article. The writer’s piece focuses on millennials and how many every-day life experiences have been affected by the internet. The author follows his quote about how real life and virtual experiences are interwoven with the findings of a Beloit College survey. Professors asked students to choose between two options: going on 1 trip to the Taj Mahal, or taking unlimited virtual trips with large pictures and interactive features. 100% of millennials chose the real life experience, juxtaposing Isaak’s claim that we can’t tell the difference. Guess he didn’t make it to the end of the article.

Regardless, Isaak published another piece in 2016 for International Snow Science Workshop. He opens claiming that social media use in the outdoors is a hotly debated topic, citing 8 different articles and papers, but none are discussing it as a dangerous influence. Arguing that it ruins the “purity” of outdoor sports? Yes. Highlighting the rise in influencer marketing that’s pulling sponsorship budgets away from award winning athletes? Absolutely. But no one else is endorsing the narrative that recreationalists are making risky decisions for the ‘gram. He goes on to quote his own 2013 paper and uses the same WIRED quote to make the argument about social proof, but extends the conversation to also include Acceptance, the A from FACETS. This concept is the tendency to make decisions that will result in approval and respect from others. I’ll give him the credit that this addendum seems way more plausible, he doesn’t have any sort of data to back up the claim. Further, out of all the FACETS introduced in McCammon’s 2004 paper, Acceptance is more extrapolative. For heuristics like the Expert Halo or Social Proof, study participants identified whether they had an identified leader or whether they ran into other parties. For Acceptance, McCammon noticed that mixed gender groups were more likely to have higher risk exposure scores than all-male groups, and extrapolated that data to mean that men were trying to impress women. No follow up questions were asked to verify that reasoning. Likewise, 2010 data also refutes it, as surveys done in Canada (McCammon’s is limited to the US) shows that men were less likely to experience an avalanche when their parties included women. It also raises the question of what gets the double tap on social media. One of Isaak’s cited sources laments the sponsorship of hybrid athlete-storytellers over legendary athletes with who have pioneered an extensive list of firsts. Some of the most liked content under hashtags like #SkiUphill and #EarnYourTurns are selfies and brightly colored stills from the skintrack. Plenty of influencers are kitted out for free, not because they’re the most intense athlete in their community, but rather because they have an interesting and different story to tell and have strong engagement from a large audience. So if content doesn’t need to be rad to be likable, could the Acceptance factor through social media potentially be protective vs. the old outdoor narrative that celebrated dominance, like first ascents/descents and speed records?

But further, Isaak’s theory isn’t supported by usage data. Below is a chart with the annual fatalities in the US, along with the trailing 5 year average plotted in red. Notice that the largest rise in avalanche fatalities occurs between 1991 and 2002, and then oscillates through the rest of the decade. Note that Facebook launched in 2004, the iPhone dropped in 2007, the Facebook app launched in 2008, Facebook album sizes increased in 2009, Instagram hit in 2010 and later rolled out to Android users in 2012. Social media adoption grew rapidly through 2016, when growth started to level out and slow down. If you consider 2000-2004 “pre-social media” years and 2012-present as full-fledged engagement, average fatalities have decreased from 27 per year to 20.

fatalities by year2

Further, this paper estimates that the backcountry use since then has increased 8-fold (a conservative estimate based on the number of users leveraging avalanche advisories). If the fatality rate stayed constant while usage was increasing, we’d be seeing up to 200 avalanche fatalities a season. Instead, we’re seeing the overall rate decrease while social media platforms are released and increase usage.

On top of that, Erich Peitzsch’s research shows that the average age of avalanche victims has increased significantly since the 1990s, and that the 30-39 age group is the only range seeing the most significant increases in the number of fatalities – not the college aged students Isaak recounted anecdotally from his classes. In recent years, fatalities of people in their 30s has outnumbered those in their 20s. However, social media usage for people in their 30s is significantly lower than those 18-29.

In addition, while McCammon’s paper and the Canadian study disagree as to whether men are safer with women in their group, they both agree that women are less likely to be involved in avalanche accidents. But guess who uses social media more often? Women.

There is not a shred of fact-based data that lends any credence to Isaak’s theory that social media increases our risks in the backcountry. And if I may add a piece of anecdotal evidence myself, I learned that avalanches were more than a model of Chevy truck through social media. I’ve shared avalanche warnings in social media groups of 40,000 women, most of whom don’t have avalanche education and answered questions about which routes have avalanche terrain and alternative options when the danger ratings weren’t low. I’ve got a stream of live updates from our local avalanche forecasters populating every day on Instagram – I don’t even have to visit the local avalanche center’s website to stay informed. I won’t make the argument that social media is keeping us safer, but it’s certainly plausible.

So why does this bother me so much? Because the narrative that accidents happen to inexperienced people pushing their luck for Instagram likes keeps us from learning from others’ mistakes. The American Alpine Institute has a wonderful piece on How to Read an Accident Report that these are stories we tell ourselves to psychologically defend ourselves from acknowledging that we participate in unsafe sports. Instead, we should try to put ourselves in the victims’ shoes and imagine what circumstances would put us in the same situations, making similar decisions. Those situations become flagged in our minds and when we see that pattern of events playing out in our real adventures, we have our “action scripts” mentally prepared to make even more conservative decisions. This imaginative process is called Research Primed Decision-making and is used in the armed forces and in emergency services to help people make good decisions in highly complicated situations. When you write victims off as narcissists risking fate for a selfie, not only are you a giant dick, you’re doing yourself an incredible disservice.

I’d like to take a moment to note that I’m not a total monster and that I reached out to Jerry Isaak privately in June of 2018 at two Plattsburgh email addresses questioning the validity of his publications and did not get a response. I may have completed my senior thesis in a 32 hour sprint in the library about 5 days before graduation, but if there’s one thing I learned in the associated lectures, it’s that academic publishing values truth over ego. Academics are expected to show their work, consider critiques, and share their reasoning if they disagree with a review. Likewise, I’m disappointed with Lynne Wolfe as editor of the Avalanche Review since 2002, for putting the original story to print in what they consider a “trade and scientific journal.” It may not be a large publication, but the content is reaching conventional publications and influencing the way the outdoor community conceptualizes accidents. In my view, the standards should be higher.

And lastly, I’d like to send condolences to Miha Sumi & John Jenkins’s friends and family for not only losing a loved one, but all of the bullshit comments from the internet peanut gallery. If the fact that the “selfie inflicted damage” narrative is untrue and detrimental to your own health isn’t reason enough to stop regurgitating these stories, remember that the outdoor community is small and that victims’ loved ones are likely reading your armchair commentary. Post accordingly.


One thought on “The Fallacy of Instagram & Alpine Accidents

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 )

Connecting to %s