I'm not boring you, am I?: Read Receipts and Intrusive Speech

Synchronous forms of communication - from speaking to someone in earshot (or signing in 'eyeshot') to telephone and video calls - have a power structure missing from asynchronous forms of communication. They restrict when the audience must (i) attend to the content of the communication and (ii) make any appropriate response. By engaging in a synchronous form of communication, the speaker is attempting to restrict the autonomy of the hearer, attempting to make them do something - to listen and respond - on the speaker's terms.¹

Of course, there can be pre-arranged and ad hoc contexts where there is consent for this, such as a meeting or a dinner date or a family meal. There are also contexts where consent has to be obtained, such as strangers on a train or calling someone unexpectedly. The limitations on contextual cues about consent in the case of telephone calls has led to the convention of the caller identifying themselves and the purpose of the call before asking 'Is this a good time?'. By and large we navigate these norms of communication well and learn strategies for dealing with abuses.

The second constraint created by the speaker - that on the audience's attention - is often hard to deal with. The audience could, of course, just ignore the speaker but that isn't always easy. Take 'ghosting' for example: while not answering the phone to an unknown number is fine, not answering to a known number is not. There seems to be an assumption in the negative judgement we make about such behaviour, embodied in the thick evaluative concept of ghosting, that consent for someone to initiate synchronous communication persists until explicitly withdrawn.² Similarly, it is very hard to say to someone already speaking to you that you want them to stop talking, that you don't want to have to hear what they are saying (at least where the content is merely boring, irrelevant or just not something you want to think about right now). Even passive 'tuning out' can provoke an angry response. This is for the obvious reason that the speaker is speaking precisely because they 'have something to say': it is important or interesting to them and they think it should be important or interesting to the audience. Learning it is neither feels like a judgement on their worth, an insult.

A case can be made that this way of thinking, that it is rude to ignore a speaker, actually gets the ethics of the situation the wrong way around. No one has a right to be heard outside specific formal and legal contexts, but surely everyone has a right to control their own thought processes, to choose what they think about and when. Let's call this the right to 'freedom of consciousness'. If someone speaks to me without my consent then they have, if only in a very minor way, infringed my right to think about whatever I want. Synchronous communication intrudes upon my thoughts, it forces my attention away from what I wanted to think about onto what the speaker wants me to think about. It is only a slight exaggeration to call this a form of verbal assault: intrusive speech, not just offensive, harassing, or threatening speech, may also be unwelcome, annoying or even distressing.

Digital Blurred Lines
Some forms of digital communication are neither clearly synchronous nor clearly asynchronous and this causes some unacknowledged problems. Email, to take the most widely used form of digital communication is clearly asynchronous. Of course, it is usually fast enough to enable a synchronous exchange, but normatively the expectations are those of asynchronous communication: the 'audience' remains free to choose when (if at all) to attend and when (if at all) to respond.³

Given that email is usually fast enough to sustain synchronous communication, how did it come to take on the normative structure of asynchronous communication? Well, one obvious factor is that email gradually took over almost all the functions of snail mail, and that is the archetype of asynchronous communication. Thus it became culturally dominant as a form of asynchronous communication.

But it is also worth noting that technically it is not very well suited to synchronous communication: the smtp protocol does not let the sender know when a message has reached the destination server, let alone when it has been downloaded or read. That is why marketing companies have had to find ways of embedding trackers into the body of an email which will do the same job.

Why does that technical restriction matter? After all, the SMS protocol is similar in this respect and yet that seems to provide for synchronous messaging.

Here we need to take care to distinguish between a technology that can be used for synchronous communication when other external conditions are in place and one which is of its nature synchronous, like the phone call. Email and SMS can be used for synchronous messaging by agreement of the two parties. They need to agree to stay online and respond promptly. Such agreements can be more or less explicit, sometimes just a temporary situation of joint attention created by the initial exchange, but they rest entirely on active co-operation. First there is consent to synchronous conversation and then email/SMS is the chosen instrument. Thus there is no possibility of non-consensual constraints on when the recipient attends and responds, because as soon as the consent is withdrawn, the conversation becomes asynchronous.

All this changes when the messaging protocol involves a presence indicator. The archetype for this is XMPP: Extensible Messaging and Presence Protocol. Created in 1999 it became the foundation - in the end conceptually rather than as a specific protocol - for all the familiar 'social media' messaging apps we have now. What such protocols allow, that email and SMS lack, is those two now ubiquitous features: Last seen and Read Receipts.

These features blur the line between an asynchronous technology and a synchronous one. If the sender knows the recipient has seen their message, and recipient knows the sender knows that, then that creates an expectation of when the recipient will respond. This is similar to the expectation that the audience will respond immediately in a face to face conversation, and flouting of the expectation creates the same implicatures.

Last Seen and Read Receipts interact in a particularly vicious way: if the recipient is online or last seen after the message was sent, but it remains unread, then it looks like 'ghosting' to the sender. But if it is read and there is no response, it implies the sender's message is uninteresting, not worth the effort of responding - like 'tuning out' while someone is talking.⁴ These technologies are built into the messaging system and they create the power structure of synchronous communication.

Opting out
Most messaging apps have privacy settings which allow users to opt out of 'Last seen/Online' and some, such as Signal and WhatsApp,⁵ also allow users to opt out of 'Read Receipts', though that is two-way: if others can't see that you read a message, nor can you see they have read a message.⁶

The first point to note about this is that by having these features on by default, opting out does not change the character of the communication technology from synchronous to asynchronous. It merely gives the recipient a strategy for dealing with the constraints on their autonomy created by the sender/speaker in a synchronous conversation.

Secondly, 'Last seen' is clearly very privacy invasive. It is hard to imagine a situation in which you would want all your contacts to know that. You might want them to know you are online now, so ready to attend to their messages, but if you are offline, why would you want them to know when you were last online? It is the sort of creepy tracking data that surveillance capitalism is so enamoured of, and I guess most users know at some level that it is being collected. But shared? This really should be off by default and that it isn't shows how the companies behind these apps want to normalise our loss of privacy.

Thirdly, the 'Online' status puts you 'in the room' for the person who wants to initiate a conversation: it tells the sender, in advance of their decision to send the message, whether the recipient is available to attend to that message. It thus creates a context in which a failure to respond almost immediately is likely to be interpreted as the recipient not being interested in what the sender has to say.

Given the scale - and complexity in terms of relationships - of most people's contacts list, that does dramatically increase our vulnerability to intrusive speech. The facility to apply some selectivity in which contacts see this information helps mitigate that effect, but it places the burden onto the potential recipient of such messages to protect their freedom of consciousness, rather than keeping the burden on the sender to consider whether the message might be intrusive.

It is important to notice that we are in a situation where the norms and expectations around synchronous messaging were originally formed in the context of communication 'technologies' where it is transparent and usually quite easy to make yourself unavailable to be an audience. We can find physical spaces away from others, we can put in our earbuds, we can put our phone on silent etc. These are easy to implement on an ad hoc basis and with the possibility of their being over-ridden for the right conversation partner. The way 'Online' status is implemented makes it fiddly to do anything on an ad hoc basis (if we still want to use the app) and demanding to set up complex filters.⁷

Of course, we can hope that the sensitivity displayed by many people in being cautious about initiating conversations when someone is 'Online' and respecting their choice about whether to attend and respond, become widespread norms in society, but the nature of the technology, the ubiquity of some messaging apps, and the barriers just discussed, make that unlikely.

There is one simple thing we can do to protect our freedom of consciousness from some of these effects: use different messaging apps for different people/groups. This is the closest we can get in a world of digital messaging to being able to choose who we sit next to in the canteen.

Read receipts
We noted above that Read Receipts can only be turned off in a minority of messaging apps.⁸ In most cases I suspect that this is a technical limitation: the protocol requires they are implemented for all users of a server or none. However, the question I am interested in is not the reason for the default / required implementation of Read Receipts but the effects they have on the normative structure of digital messaging.

In face to face conversations we have various ways of indicating that we have 'read' what is being said to us before we are expected to respond: eye-contact, facial expressions, nods and shakes of the head, subvocal noises etc. Failure to indicate attention like this can lead the speaker to question whether the audience is attending and possibly take offence that they are not: "I'm not boring you, am I?"

Sending these signals that one has heard what someone is saying is under the voluntary control of the audience, and in some cases - when there can be space for doubt that the speaker was heard by the audience - we can protect our freedom of consciousness by pretending we haven't heard and not sending those signals. That space of uncertainty creates an important space of autonomy where the audience can choose not to be part of the conversation without giving offence..

Read Receipts create no such space because the message recipient has no control about whether they are sent. The fact that the internet is full of tips on how to get around this problem - such as turning on 'Airplane mode' to read the message - shows that it really is a problem, that people find their lack of control on a case by case basis over whether Read Receipts are sent to be an infringement of their autonomy worth some effort to avoid. Part of this is the fact that messaging is, as I noted above, intrinsically a medium of synchronous communication so has the capacity to be used for intrusive speech.

Read Receipts increase the capacity for intrusive speech by removing one additional layer of control from the audience: they create a situation which is - conversationally - like having everyone who has your contact details sitting across the room from you all the time. Any time they choose to speak, you become constrained in when you must respond (pretty much as soon as you read the message) and thus have no freedom to choose. But unlike actually sitting in a room with someone, you have no control over whether you are in the same room as that person at that time. Furthermore, they cannot see if you are reading a book or listening to music or otherwise exercising your freedom of consciousness, so cannot hold back from speaking until a time when they know you may wish to hear what they have to say.⁹

Be Gentle
Sometimes we really do need to know that someone has received a message, because it is time-sensitive and if they haven't we may want to find another way of contacting them or another person to contact. And Read Receipts do this very well. Of course, the digital equivalents of smiling, nodding etc. also do that and it is possible the prevalence of Read Receipts makes these less used (though 'Reactions' are very popular precisely because they replace a lot of face to face nonverbal communication).

So I would advise everyone to turn off Read Receipts where they can and start to practice active - rather than passive - acknowledgement that the message has been received. Take control where you can. Use different messaging apps for different people/groups. Encourage others to do the same. Try to acknowledge that freedom of consciousness is a thing and it matters, but most of all that the digital world draws us into behaviours that chip away at that space of autonomy. If we don't actively resist, we will lose yet another freedom to Big Tech.[^9]

It is also really important to recognise the possibility that your message might constitute intrusive speech for the recipient. Just like you don't know what someone is doing - or thinking - when you call them, so you start by checking the call is welcome, remember that that you don't know what they are doing or thinking when you message them. Their priorities may be different from yours, they may not want to attend to and respond to your message right now. But they may need to read it to make that assessment. So be gentle on the recipients of your messages. Allow them to take their time. Or even save your message for when you have reason to believe they will be receptive, until you have the virtual equivalent of being in the same room.

[^9] There is a lot more to be said here about the attention economy shaping a culture of always-on digital life through fear of missing out and the dopamine hits of peer approvals. But that is a different topic.