Monday, December 23, 2013

Why Don't We Host Strangers Anymore?

Note: This is an old post from October 2013, originally at

When my grandpa was in high school (1950’s), he liked to spend long weekends going on hitchhiking excursions with his friends. They would hitch rides along the highway and sleep on the porches of friendly strangers in small towns, working their way from outside Detroit to cities hundreds of miles away. Sometimes when they couldn’t find a place to stay, they asked police for directions to good public places for sleeping outside.

Darara tells a similar story about his mom’s experience growing up in Ethiopia. She would walk across the countryside to her school which was far away, and if she couldn’t make it home before dark, would often stop at the houses of strangers, who were happy to feed her and offer her a place to sleep.

One more case study - when Charlie’s dad was a 20-something working for the Peace Corps in Borneo, he and his coworkers spent their weeks off hiking between local villages. They travelled with few supplies, because at each village they arrived in, they - perhaps something of a novelty as foreigners - were welcomed, fed, and given places to sleep.

Needless to say, I cannot travel around the US (at least on the East Coast) today like my grandpa once did. Why are some people in space-time so willing to do things for strangers? Why are they unwilling where I live?

For the purpose of investigation, I’ll limit the discussion to a specific case - that of hosting a not-dangerous-looking stranger in one’s house for one night. Now we ask the following questions:
  1. What factors lead / led people to host strangers? 
  2. What factors led to this change in the US between the 1950’s and now? 
  3. How good or bad is it that this change occurred? 
  4. How might things proceed from their current state? 

In what follows, I’ll describe the thoughts Darara and I had on these questions at lunch on Friday and Saturday. These ideas are based on our own experiences and stories we’ve heard, but not on any research or data - so they could be very wrong indeed! Please contribute your own thoughts - what did we get right or wrong? What did we miss?

1. What factors lead / led people to host strangers?

Strangers might be dangerous - maybe they are thieves, murderers, or rapists. Given this possibility, why would one let them into one’s home? We thought of two reasons:

(i) One identifies with the hostee, for whatever reason.

Darara’s mom’s hosts were of her same ethnicity, cultural and religious background, and larger community. That being the case, they likely identified as similar to her - making them (a) less likely to be afraid that she was dangerous because they perceived themselves as not dangerous, and (b) more inclined to be nice, in the way that we are all especially nice to someone who we learn grew up in our hometown, or shares the same religious beliefs, or in some other way shares our experience - we like people similar to ourselves. It’s sort of like her hosts were hosting a 2nd cousin.

Similarly for my grandpa. I’m confident in the speculation that people would have been less friendly to him as, say, a black teenager in the 1950’s. But he probably came from the same white, middle class, child-or-grandchild-of-immigrants background that many of his hosts did. So they projected perceptions of their own kids or neighbors onto him, and felt like he was a part of their greater community.

Identification with the hostee explains some cases of hosting. But certainly Charlie’s dad had little in common with those who were so welcoming to him. So I think there is another significant factor.

(ii) The social expectation that one will host people.

In this regime, the host doesn’t have to think it’s a smart gamble to host the stranger - rather, they just do it because it is just a thing that one is expected to do.

This is best illustrated by another example: If I know that there is a Harvard pre-frosh who needs a place to stay, I will almost certainly be willing to host him or her. On some level, I understand that this stranger might steal my laptop or pee on my couch, and I expect to gain little from the interaction. One might say I host altruistically, knowing that it’s worth a small amount of trouble and risk on my part to give this person a place to stay - but they could almost surely find another host otherwise, so I don’t really face that tradeoff. Rather, I think the reason I’m willing to host pre-frosh is just that it’s expected that I will do it. It’s not as much of a decision as a conformity to a social rule.

This could help explain the experiences of Charlie’s dad.

2. What factors led to this change in the US between the 1950’s and now?

(i) Mass media.

No doubt, some percent of the hostees have always been thieves, murders, rapists, etc., but the percentage seems to be pretty low. So chances are, no one you know has ever had a malicious hostee. If those are the only people you acquire information from, then you probably think hosting is safer than it is. And if one of the few people you know has had a malicious hostee, then you probably think hosting is more dangerous than it is. Many people are misinformed, but on average, they have roughly accurate expectations of stranger-danger.

Now we introduce the 21st century media - a flurry of internet gossip that favors the most astonishing stories, independent of their distance from the reader. This leads to sample bias towards extraordinary events, which, in the case of hosting strangers, are mostly malicious hosts and hostees. So now we hear a few mild but nice stories from our friends and neighbors about the nice stranger they hosted, and also an alarming amount of terrifying and tragic stories about the worst cases from across the country.

The result: a paranoia that (a) makes us less likely to identify with strangers and instead overestimate the chances of stranger-danger, and (b) a change in social norms that makes it acceptable or even expected to turn away strangers.

(ii) Distant social circles.

Before the the popularity of airplanes, interstate highways, the internet, and telephones, people mostly interacted with those in physical proximity. This made it important that your in-person interactions went well, because (a) you didn’t have any other interactions and (b) there were fewer people you could interact with in general, so each interaction mattered more. This conditioned people to think their in-person interactions really mattered - and so they were nicer to the stranger on the doorstep.

Contrast with today - Americans spend a significant amount of time traveling to distant relatives, talking on the phone, Skype-ing, and getting lost in TV shows or other virtual worlds. As a result, we’re less conditioned to care about how well a given in-person reaction goes - and more likely to turn away a stranger on the doorstep. We can go make some more “friends” on the internet instead.

(iii) Decreased cultural homogeneity.

Note: my evidence for this point is mostly impressionistic and could totally be wrong, or biased my place is in society:

I get the impression that - compared to two random Americans from the 1950's - two random Americans in the US today, have, on average, less in common - fewer shared experiences, different assumptions about the way the world works, less consensus on who really matters, etc.

This is certainly true with respect to economic diversity. The spread of wealth in the US is wider than in the 50’s, and the idea of the American middle class seems to be disappearing. Sociopolitically, the end of WWII could have also been a unifying factor in the 50’s, and there was a lot more pressure (think McCarthyism) not to deviate from social expectations.

The internet (ala point (ii)) also contributes here. Since we can now find people who share even our most arcane interests, there’s less pressure to conform to norms. We don’t get along with our physically local community, but we can find an online community of people like ourselves instead.

Finally, our mass-media-induced paranoia (ala point (i)) inclines us to feel that people are different from us, whether we are or not.

3. How good or bad is it that this change occurred?

Reasons why it might be bad:
  • I can’t go on cool adventures like my grandpa did. 
  • My stress level is generally higher, because I don’t trust the people around me. I feel calm and secure when within communities of people I trust and love (for me it’s my family, or Studio Band, or Bredline, etc.) 
Reasons why it might be good:
  • I avoid being taken advantage of by malicious hosts or hostees. 

So where are we?

I think we are at least slightly too afraid. The information on which I would base my decision to host or not host a stranger is biased heavily by the media, so I probably overestimate the chance of having a malicious hostee and act too cautiously.

That said, this is probably better than the super-naive historical alternative: suppose we each were to calculate our chances of stranger-danger using only the experiences of people we personally know. Since we’d each have a small sample size, most of us would expect that hosting is perfectly safe - safe enough that we’d do it - and a few of us would expect it’s much more dangerous than it is - dangerous enough that we’d abstain. The point is that we’ll each either make a careless decision or a needlessly paranoid decision, so outcomes should be suboptimal, probably on the side of too little caution.

4. How might things proceed from their current state?

(i) We can’t go back not the way we came.

Right now hosting and being hosted are both pretty heavily stigmatized. This means that there are roughly two types of hostees: those who are exceptionally friendly and don’t mind the stigmas, and the malicious ones who are trying to take advantage of hosts. So if I suddenly have a change of heart and announce I’m willing to host strangers, I’ll mostly get people from these two groups - and not from the much larger third group of non-malicious people who were scared away by the stigma. This means that at least in the short term, I’m going to get taken advantage of for being nice - so there’s a big disincentive for me become a more willing host.

You could say we’re stuck at a local maximum - it might be better for everyone if hosts and hostees were more willing, but the first few to try it will get screwed.

(ii) Likely things will be stuck this way.

(iii) Maybe certain internet communities can change things.

In 2003, a website called Couchsurfing started enabling strangers to host each other. My understanding is you join the website and can either offer to be a host, or search for people who will host you. The site attempts some security measures by encouraging hosts to ask for references and read reviews from past hosts, and it looks like so far, the network has appears to have been quite safe. Some noise ( on the internet suggests that it may not be so safe, or that its safety only comes from that the network is a very small and self-selecting subset of the general public. Either way, it’s a cool phenomenon, and maybe a good way to bring back the idea of hosting strangers.

Internet communities have a broader potential. Certain communities, like Reddit (and especially certain subreddits I think), seem pretty socially cohesive. Redditors tend to share many social views and some interests, and are somewhat accountable to each other through their usernames, which for some serve as coherent identities. Reddit even has mini-holidays where people mail gifts to each other or meet in person - which represents some amount of actual trust. It’s conceivable to me that redditors would be willing to host random other redditors, just because they identify with each other. It would also be easy to do some background checking on such a person’s Reddit history, which might make someone feel safer.

But why are internet communities any different from other communities? Why wouldn’t they suffer from the same problems?
  • Mass media doesn’t classify people (yet) based on which internet communities they belong to, though they do classify people by race, religion, state, etc. This would enable, say, a redditor to believe that people in the US are generally dangerous, but their Reddit community is a small and safe subset. Also, sites like Reddit might try to do their own reporting of malicious incidents between redditors, but in a more statistics-presenting and less fear-mongering way. 
  • The distant social circle issue doesn’t apply here; internet communities are part of our new distant social circles. 
  • Diversity - internet communities tend to be smaller and more unified by interest than geographic communities, so there’s more association between members. 

It’s important to note that the sort of hosting that would come from these internet communities is NOT as all-encompassing as the vision I have of the 50’s. It’s not that you can just go to a random door, but that you can look up all the doors of those in your online community and pick one of those that’s nearby. The internet gives us a way to find the doors of people most similar to us, and within that framework, it could offer better accountability.

That’s all for now - thanks for reading the speculations Darara and I have been making! Let us know what sounds right, what sounds wrong.

No comments:

Post a Comment