I am often skeptical of arguments about China that rely heavily on personal experience. The author may explicitly acknowledge that their experience is limited without appropriately treating their arguments with rigor beforehand. This acknowledgment may thus sit beside conclusions about the values, behaviors, and functions of Chinese society without meaningfully constraining any of them.
Different speakers may each have peers with similar experiences that corroborate their claims, or repeated experiences that solidify their firsthand evidence. We may believe one or another speaker is more perceptive, well-informed, or more willing to acknowledge facts contrary to their beliefs. However, the critical question is how far these accounts support the conclusions being drawn.
I will address here how we should evaluate firsthand accounts, including when people offer conflicting accounts.
The Mistake of Universalizing Personal Experience
Happy two-year anniversary to this tweet by culture critic Rayne Fisher-Quann.
When people use personal experiences to support arguments about a locale, collective, or institution, one must best establish which conditions each person has encountered and how widely those conditions apply. An experience tells us something happens, reveals how an institution works, or challenges a claim about what never happens, etc.. Claiming that this experience describes a broader population requires a further argument about how widespread it is, which conditions produce it, and whether the people we encountered were especially likely to experience it.[1]
Having lived experience indeed provides substantial tacit and nonpublic knowledge, but proximity, salience, attention, and memory all affect the sample of observations. This knowledge may be most directly applicable to a narrow environment whose relationship to the broader population remains unclear.
The same problem can arise in impressions of the US. Suppose an international student’s acquaintances are concentrated in a well-resourced university and its professional networks. These environments are insulated from the experiences of the average American and do not generalize well; conclusions about the American psyche and economic opportunities in the US would require further arguments apart from personal experience in these networks. Furthermore, embedding oneself for additional years among similarly situated people does not automatically lessen this asymmetry, even if it substantially improves one’s understanding of that environment.
I have caught myself making similar generalizations about the Chinese internet. Mainland Chinese platforms harbor cynical, absurdist humor, as well as slop content and pop social commentary. Moreover, I encounter plenty of misanthropic engagement bait and comment section debates. However, these experiences cannot establish that this behavior is more prevalent in China.
Certainly, I’ve witnessed a great deal of slop content and engagement bait on Western platforms. I’ve also witnessed altruistic behavior and crowdfunded philanthropy in both ecosystems. These behaviors may be expressed in different forms on different topics. Their relative visibility to me, however, cannot settle how prevalent they are or what produces them.
Of course, my argument is not that comparisons are only valid if they hold every other condition constant. Rather, when a question concerns the average conditions or general attitudes across two countries, one’s differential exposure to social experiences is often a part of the answer underestimated or unaccounted for.
Standpoint Epistemology
Sometimes, one further argues that their social position privileges them with an epistemic advantage, and thus grants authority to their insufficiently supported inference. This account is thus perceived as more convincing by the audience, which often does not have the contextual fluency to offer it the appropriate evidential weight. This is similar to standpoint epistemology in recent practice.
Certainly, taking relevant experience as a reason for provisional trust is different from taking it as sufficient reason to settle a broad disagreement. We consistently rely on imperfect indicators of competence, and one's language and contextual fluency can reasonably affect how much weight we give an account. Often, uniform information gathering is difficult, and we use these subjects as proxies representative of a demographic. Their experiences need not be representative for their knowledge to be useful.
However, the people available to speak for a group in often influential spaces may already be systematically different from those they are meant to represent.[2] More concretely, access to English-speaking AI safety circles may favor particular educational and professional backgrounds. These backgrounds may provide substantial knowledge of particular institutions and networks, while leaving other environments less familiar.
These individuals are not unqualified to speak about China, of course. When it comes to broad claims of national character, however, the experience they have may not support a general claim about a people and their circumstances.
These experiences may be corroborated by others in similar positions, but have limited transferability to the experiences of the average China resident, Chinese national, member of the Chinese diaspora, etc.. Several people may accurately describe similar experiences and still leave much of the relevant population unaccounted for. As such, conclusions about the Chinese psyche or national character, as well as broad social norms, requires more rigor than agreement of people whose experiences overlap.
Countering Generalization in Practice
It would, of course, be imprudent to shun standpoint theory altogether, as social position indeed structurally affects what people encounter and what they need to understand. Research literature, moreover, is not a sufficient substitute for lived experience, as researchers themselves often commit blunders when scoping questions, sample groups, and measurements of the phenomenon they are attempting to isolate.
As one example, early planning for the Nam Theun 2 dam in Laos underestimated how many people depended on the downstream Xe Bang Fai river system, particularly failing to consider those who lived away from the riverbanks. Later interviews with villagers also revealed that efforts to replace lost livelihoods were insufficient. Supplied chickens died of disease or were sold because feed was too expensive. Direct payments addressed only one or two years of income. It is unclear these oversights can be remedied without inquiring those affected, and those conventionally excluded from inquiry.
My objection concerns, rather, the extension from this privileged evidence to the conclusions drawn from it. One using individual experiences as evidential backing for a broad claim should additionally consider the external conditions, sequence of events, etc.. More importantly, one should strive to disprove the alternative explanations and give reasons to prefer the proposed explanation.
Functionally, this is to say: one should ask oneself what else could explain what one observed and whether evidence beyond one’s immediate environment supports it. This includes reading statistics and studies, or seeking out those whose circumstances differ from ours.[3] One should then explicate this process in writing to reflect the limited applicability of one's experiences.
Obviously, this is not meant to exert a chilling effect on the discourse and have individuals resort to self-censorship out of apprehension or exhaustion toward a demanding burden of proof. Rather, these additions would make discourse more productive and preempt many back-and-forths surrounding refuting explanations and additional circumstantial information regarding the subject and their experiences. At the very least, one would be more likely to turn away from overreaching claims and toward building a holistic model of China if one interrogated their experiences beforehand.
Pierre Bourdieu elucidates this much more clearly in “Preface to the English-Language Edition,”Distinction. Though his argument circles around the limited applicability of research grounded in French society to describing phenomena in American society.
The internet has made it easier to form friendships with people I would probably never have met otherwise. Over time, ordinary conversations about work, family, and what we want from life have changed some of my assumptions about China. These friendships have also made me more cautious about treating my own experiences or those of my peers as widely shared.
Relying too much on personal experience is bad, but arguing against a strawman position you did not take, believing you have a good understanding without any personal experience is bad too.
For example, until I went to China, I didn't understand China was big. I know this sounds impossible. If you had asked me before I went "do you understand that China is big?" I would have said "yes of course I understand it's big, I'm a well-traveled adult who does fermi estimates for fun, not a five-year-old". Still, until I flew into a third-tier city I've never heard of in China, and saw the 90 metro station and 40 elevated transit systems and highways of electric cars, I did not understand China was big. And there are multiple nuances about China more subtle than "it's big!" that I just would not get if I hadn't been there.
I am often skeptical of arguments about China that rely heavily on personal experience. The author may explicitly acknowledge that their experience is limited without appropriately treating their arguments with rigor beforehand. This acknowledgment may thus sit beside conclusions about the values, behaviors, and functions of Chinese society without meaningfully constraining any of them.
Different speakers may each have peers with similar experiences that corroborate their claims, or repeated experiences that solidify their firsthand evidence. We may believe one or another speaker is more perceptive, well-informed, or more willing to acknowledge facts contrary to their beliefs. However, the critical question is how far these accounts support the conclusions being drawn.
I will address here how we should evaluate firsthand accounts, including when people offer conflicting accounts.
The Mistake of Universalizing Personal Experience
Happy two-year anniversary to this tweet by culture critic Rayne Fisher-Quann.
When people use personal experiences to support arguments about a locale, collective, or institution, one must best establish which conditions each person has encountered and how widely those conditions apply. An experience tells us something happens, reveals how an institution works, or challenges a claim about what never happens, etc.. Claiming that this experience describes a broader population requires a further argument about how widespread it is, which conditions produce it, and whether the people we encountered were especially likely to experience it.[1]
Having lived experience indeed provides substantial tacit and nonpublic knowledge, but proximity, salience, attention, and memory all affect the sample of observations. This knowledge may be most directly applicable to a narrow environment whose relationship to the broader population remains unclear.
The same problem can arise in impressions of the US. Suppose an international student’s acquaintances are concentrated in a well-resourced university and its professional networks. These environments are insulated from the experiences of the average American and do not generalize well; conclusions about the American psyche and economic opportunities in the US would require further arguments apart from personal experience in these networks. Furthermore, embedding oneself for additional years among similarly situated people does not automatically lessen this asymmetry, even if it substantially improves one’s understanding of that environment.
I have caught myself making similar generalizations about the Chinese internet. Mainland Chinese platforms harbor cynical, absurdist humor, as well as slop content and pop social commentary. Moreover, I encounter plenty of misanthropic engagement bait and comment section debates. However, these experiences cannot establish that this behavior is more prevalent in China.
Certainly, I’ve witnessed a great deal of slop content and engagement bait on Western platforms. I’ve also witnessed altruistic behavior and crowdfunded philanthropy in both ecosystems. These behaviors may be expressed in different forms on different topics. Their relative visibility to me, however, cannot settle how prevalent they are or what produces them.
Of course, my argument is not that comparisons are only valid if they hold every other condition constant. Rather, when a question concerns the average conditions or general attitudes across two countries, one’s differential exposure to social experiences is often a part of the answer underestimated or unaccounted for.
Standpoint Epistemology
Sometimes, one further argues that their social position privileges them with an epistemic advantage, and thus grants authority to their insufficiently supported inference. This account is thus perceived as more convincing by the audience, which often does not have the contextual fluency to offer it the appropriate evidential weight. This is similar to standpoint epistemology in recent practice.
Certainly, taking relevant experience as a reason for provisional trust is different from taking it as sufficient reason to settle a broad disagreement. We consistently rely on imperfect indicators of competence, and one's language and contextual fluency can reasonably affect how much weight we give an account. Often, uniform information gathering is difficult, and we use these subjects as proxies representative of a demographic. Their experiences need not be representative for their knowledge to be useful.
However, the people available to speak for a group in often influential spaces may already be systematically different from those they are meant to represent.[2] More concretely, access to English-speaking AI safety circles may favor particular educational and professional backgrounds. These backgrounds may provide substantial knowledge of particular institutions and networks, while leaving other environments less familiar.
These individuals are not unqualified to speak about China, of course. When it comes to broad claims of national character, however, the experience they have may not support a general claim about a people and their circumstances.
These experiences may be corroborated by others in similar positions, but have limited transferability to the experiences of the average China resident, Chinese national, member of the Chinese diaspora, etc.. Several people may accurately describe similar experiences and still leave much of the relevant population unaccounted for. As such, conclusions about the Chinese psyche or national character, as well as broad social norms, requires more rigor than agreement of people whose experiences overlap.
Countering Generalization in Practice
It would, of course, be imprudent to shun standpoint theory altogether, as social position indeed structurally affects what people encounter and what they need to understand. Research literature, moreover, is not a sufficient substitute for lived experience, as researchers themselves often commit blunders when scoping questions, sample groups, and measurements of the phenomenon they are attempting to isolate.
As one example, early planning for the Nam Theun 2 dam in Laos underestimated how many people depended on the downstream Xe Bang Fai river system, particularly failing to consider those who lived away from the riverbanks. Later interviews with villagers also revealed that efforts to replace lost livelihoods were insufficient. Supplied chickens died of disease or were sold because feed was too expensive. Direct payments addressed only one or two years of income. It is unclear these oversights can be remedied without inquiring those affected, and those conventionally excluded from inquiry.
My objection concerns, rather, the extension from this privileged evidence to the conclusions drawn from it. One using individual experiences as evidential backing for a broad claim should additionally consider the external conditions, sequence of events, etc.. More importantly, one should strive to disprove the alternative explanations and give reasons to prefer the proposed explanation.
Functionally, this is to say: one should ask oneself what else could explain what one observed and whether evidence beyond one’s immediate environment supports it. This includes reading statistics and studies, or seeking out those whose circumstances differ from ours.[3] One should then explicate this process in writing to reflect the limited applicability of one's experiences.
Obviously, this is not meant to exert a chilling effect on the discourse and have individuals resort to self-censorship out of apprehension or exhaustion toward a demanding burden of proof. Rather, these additions would make discourse more productive and preempt many back-and-forths surrounding refuting explanations and additional circumstantial information regarding the subject and their experiences. At the very least, one would be more likely to turn away from overreaching claims and toward building a holistic model of China if one interrogated their experiences beforehand.
Pierre Bourdieu elucidates this much more clearly in “Preface to the English-Language Edition,” Distinction. Though his argument circles around the limited applicability of research grounded in French society to describing phenomena in American society.
Olúfẹ́mi Táíwò’s “Being-in-the-Room Privilege” (2020) expands on this argument.
The internet has made it easier to form friendships with people I would probably never have met otherwise. Over time, ordinary conversations about work, family, and what we want from life have changed some of my assumptions about China. These friendships have also made me more cautious about treating my own experiences or those of my peers as widely shared.