The politics of pronouns

A chapter from Neuroscience for Organizational Communication: A Guide for Communicators and Leaders, by Chartered Psychologist Dr Laura McHale, courtesy of Palgrave Macmillan.

In this exclusive extract Laura McHale examines pronoun use in the political speech. She explores the differences between the inclusive and exclusive “we,” and the perils of the overuse of “I.” Modal words are discussed, both strong and weak (the latter sometimes called weasel words). The emerging field of data mining and sentiment analysis is reviewed. Lastly, do we need a more authentic language of vulnerability in organisational life?

Analyzing political speech is a worthwhile way to study pronouns, especially when we understand how they work from a neuroscientific perspective. Pronoun use can be both subtle and profound so it’s worth exploring the topic more deeply, especially since it’s such a rich and relevant area for professional communication.

There is an impressive body of research into how politicians use pronouns, which is usually in a strategic way to show power, solidarity, or authority (Bull & Fetzer, 2006; Pennycook, 1994). For example, in Brozin’s (2010) analysis of a Barack Obama speech on race, he found that Obama displayed a preference for “we” pronouns to position himself as a spokesperson for the nation. He tended to use “I” pronouns in more mitigated phrases, such as “I believe” and “I cannot,” which signaled, respectively, more or less personal involvement.

Pronouns in political speech are commonly used by politicians to construct favorable images of themselves and others (Bramley, 2001). Pronouns can be cleverly deployed to show affiliation or create distance between people. Bramley (2001) shrewdly observed that pronouns are used to socially construct the identity of the politician and others, rather than objectively represent them. This is a remarkably insightful point, because it emphasizes how political realities are co-created and collectively maintained through language choices.

The Nationalist “We”

Pronoun use is also strongly influenced by culture and political orientation. For example, “we” pronouns are used extensively, sometimes exclusively, in mainland China and Communist Party rhetoric, usually to frame discourse within a nationalist context (Karlsson, 2017).

Let’s take an example of a speech delivered by Xi Jinping to the World Economic Forum in January 2021. Here is an excerpt:

The right approach is to act on the vision of a community with a shared future for mankind. We should uphold the common values of humanity, i.e. peace, development, equity, justice, democracy and freedom, rise above ideological prejudice, make the mechanisms, principles and policies of ourcooperation as open and inclusive as possible, and jointly safeguard world peace and stability. We should build an open world economy, uphold the multilateral trading regime, discard discriminatory and exclusionary standards, rules and systems, and take down barriers to trade, investment andtechnological exchanges. We should strengthen the G20 as the premier forum for global economic governance, engage in closer macroeconomic policy coordination, and keep the global industrial and supply chains stable and open. We should ensure the sound operation of the global financial system, promote structural reform and expand global aggregate demand in an effort to strive for higher quality and stronger resilience in global economic development. (Xi, 2021)

In this example, there are four instances of “we.” I did a quick analysis of the full speech (it is approximately 3000 words – a little shorter than the average chapter in this book). In the full speech there were 47 instances of “we,” four instances of “they” and four instances of “us.” There was nary an “I” pronoun in the entire speech. According to Karlsson (2017), this is not unusual for Chinese political speeches.

Just to see what would happen, in a very non-scientific experiment, I swapped out the “we” pronouns for “I.” and have indicated the changes in bold text below:

The right approach is to act on the vision of a community with a shared future for mankind. I should uphold the common values of humanity, i.e. peace, development, equity, justice, democracy and freedom, rise above ideological prejudice, make the mechanisms, principles and policies of our cooperation as open and inclusive as possible, and jointly safeguard world peace and stability. I should build an open world economy, uphold the multilateral trading regime, discard discriminatory and exclusionary standards, rules and systems, and take down barriers to trade, investment and technological exchanges. I should strengthen the G20 as the premier forum for global economic governance, engage in closer macroeconomic policy coordination, and keep the global industrial and supply chains stable and open. I should ensure the sound operation of the global financial system, promote structural reform and expand global aggregate demand in an effort to strive for higher quality and stronger resilience in global economic development.

It is striking how much the use of “I” pronouns changes the emotional tenor of the speech. The pronoun “I” can suggest a sort of manicgrandiosity that was not present in the original speech, e.g., I don’t think any sane politician would ever say “I should build an open world economy”. But in other cases, the use of “I” conveys a sense of personal responsibility and accountability that is reassuring, e.g., “I should uphold the common values of humanity”. This is instructive, because it reveals how pronoun use conveys – and sometimes reveals, other times obscures – a leader’s sense of self in the systemic context.

In the research on political language, the reliance on “we” is even more illuminating for the ways it can manipulate and what it can obscure. Often, it implies consensus or a collective mandate where none may actually exist. The use of “we” can be a method by which leaders seek to gain legitimacy.

But “we” pronouns are even more complex than that because there are really two different kinds of “we” in speech, which we will review in the next section.

Inclusive Versus Exclusive “We”

There are two different aspects of “we” pronouns: inclusive or exclusive. This concept is referred to as clusivity, which references whether the speaker is including the audience (the receiver of the communication) and others in the purview of the “we” (Filimonova, 2005).
An example of the exclusive “we” might be, “We the Communications team.” It is used to distinguish one group from another. But the inclusive “we” is meant to reference pretty much everyone (whether this is actually true or not). A famous example from my own country is, “We the people.”

It gets a little confusing because the “we” can be inclusive in some respects and then exclusive in others. For example, being Scottish in Scotland is an inclusive, “We are Scottish,” but in a political discussion, it might become an exclusive “we” to distinguish it from other parts of the UK.

The inclusive “we,” is quite expansive and usually signals broad commonalities of experience. How it is used can really vary, depending on the optics; it can appear rather brazen to speak on behalf of everyone, at least in non-totalitarian contexts. More typically, instead of the inclusive “we,” communicators will reference universal ideas and principles, such as “mental health” or “justice” or “humanity.” The inclusive “we” is increasingly seen in the language around climate change, e.g., “We must do something to address climate change”, meaning everyone must. This sense of inclusiveness is also implied as a core principle in many forms of jurisprudence, including such legal concepts as crimes against humanity and hostis humani generis (for more on these, please see Green, 2008).

In political speech, the inclusive or exclusive “we” is usually made explicit through the context. But not always. Sometimes it’s left a little vague on purpose. Who is the “we” we are talking about? It’s also interesting when politicians abandon the exclusive “we” and shift to the third person; “folks” is a common one in American political rhetoric, or even more euphemistic choices, such as “Second Amendment people.” This is a particularly crafty way of acknowledging, and sometimes tacitly nodding to, specific groups without explicitly signaling membership within them. More adaptively, it can be used to signal that there is dissent or a different opinion.

The wishy-washy “we” (have I coined an alliteration?) also happens in organizations, with surprising frequency. Leaders sometimes obliquely reinforce the status and privileges of certain groups within organizations, while neglecting others (usually without realizing they are doing it). This commonly happens in a racial/cultural or gender sense, especially because organizations tend to mirror the larger inequalities that exist in the societies in which they are embedded. But it also involves internal politics and the often unacknowledged status and privilege connected to job type and department. Organizations have high status and low- status roles, and many in between. Sometimes, leaders (and employees) talk about core businesses using an exclusive “we”, but infrastructure functions such as Compliance, HR, Risk, and Communications become “they” or “them” – unless the leader switches back to a more inclusive “we.” This type of pronoun use conveys subtle but potent messages about belonging and what teams have a legitimate stake in an organization’s success.

Why is this important for communicators? For a few reasons. Status and Relatedness are stress/reward triggers and Attachment is a core human need. The exclusive “we” may reward Status for the ingroup members, but it threatens Status for the outgroup – especially those that want to belong. For outgroup members, the exclusive “we” meddles with the sense of Relatedness and creates insecure Attachment. Those not included in the “we” (an experience common for generations of non-White Americans, for example), have a sense of not belonging or feeling welcome. This activates all the pain centers of the brain. It is a pernicious influence in our organizations and can frustrate everything from our strategies to create more creative and collaborative cultures to diversity and inclusion initiatives.

That said, the exclusive “we” is always going to be prevalent in organizational communications. After all, these are messages on behalf of a group of people working in concert, engaged in a collective endeavor. External communications and Marketing professionals rely on it: “We are committed to delivering the best solutions” or “We take your feedback seriously.”

But for more internally focused communications, we need to be particularly mindful of how the use of “we” can impact messaging, particularly for leadership communication. As we have seen, pronouns can be used to strengthen or disrupt the relationships between CEOs and employees (Warnick, 2010). They can also serve to disenfranchise people, functions, and roles.

The Perils of “I”

Now let’s turn to the use of “I” pronouns, which are equally fascinating. Just as overuse of “we” can pose certain risks for organizations, the overuse of “I” is particularly revealing for leadership communications.

Both within organizations and in the world at large, the words of politicians, central bankers, and CEOs are carefully, even painstakingly, parsed for clues around the person’s intent, personality, and leadership style.

There is a growing body of evidence that the heavy use of “I” pronouns in leadership communications is linked to narcissism and other maladaptive leadership syndromes. For example, one study explored the link between CEO language and hubristic leadership, particularly as associated with unethical and destructive behaviors (Akstinaite et al., 2019). Another study explored the ratio of “I” to “we” pronouns in CEO communication as an effective proxy for CEO narcissism (Aabo et al., 2020). And a third study focused on the size of CEO signatures and their relationship to narcissism and poor financial performance (Ham et al., 2018).

The heavy use of “I” pronouns can signal that something is seriously amiss with an organization’s leadership. Often it signals a more heroic mode of leadership, rather than viewing leadership as a distributed function (Heifetz et al., 2009). With an over-reliance on “I” pronouns, there can be a pervasive sense of unease, uncertainty, and fear throughout the organization. These are powerful stress triggers for the brain. Obviously, in these types of toxic environments, the communications themselves may be the least unsettling aspects of working there, even as we see how communication can serve as a barometer of the relative health of leadership and the long-term stability of an organization.

Strong Modals and Weasel Words

It’s not just pronoun use that is fascinating in this way. There is also some terrific research on the use of modal words in organizational communications. I was surprised to see that most of this research is published in accounting and finance journals and not those related to communication, leadership, or organizational psychology. Which is all the more reason that it’s worthwhile for communicators to take a multidisciplinary approach.

A modal is a grammatical term used to describe a word that is used with a verb to express possibility or intention. Modal words are usually described as strong or weak, depending on the degree of certainty that they signal. For example, weak modal words are terms such as may, might, could, depending, possibly, and appears, and signal less certainty than strong modal words such as can, will, shall, and are. Weak modal words are also known, more colloquially, as weasel words (Loughran & McDonald, 2016).

Studies have shown that the use of modal words in corporate communications can be an important measure of a company’s financial health. For example, one study of publicly traded American firms showed that companies with a high proportion of weak modal words in their annual reports and/or IPO prospectuses showed higher subsequent stock return volatility than firms that did not (Loughran & McDonald, 2016). This insight makes a compelling case for textual analysis as part of an investment strategy.

Organizations understandably tend to downplay negative news in their formal communications, and often carefully couch it among other, more positive messages. The problem with this approach is that negative information is so padded with positive words that the overall meaning becomes obscure. This is a big problem for investors and financial analysts alike. To solve this conundrum, two Finance professors at the University of Notre Dame, Tim Loughran and Bill McDonald, conducted a sweeping textual analysis of American corporate 10-Ks (e.g., annual reports) over a ten-year period, to see if they could ferret out ways that companies employed avoidance strategies in their communications. They evaluated expressions of sentiment and created six different (and often overlapping) word lists to categorize them: Negative, Positive, Uncertainty, Litigious, Strong modal, and Weak modal. Their overall goal was to identify words that signal avoidance strategies in communication style (Loughran & McDonald, 2011).

The resulting Loughran and McDonald (LM) sentiment lexicon is extensive. The biggest lists are of the Positive and Negative words; they found over 300 Positive and 2300 Negative words. Given that ratio, it is plain to see how many organizations will take great pains to invent countless ways of describing negative events, often at the expense of simple and straightforward language.

By the way, the ten most frequently occurring LM Negative words are: loss, losses, claims, impairment, against, adverse, restated, adversely, restructuring, and litigation. These ten words represent less than 1% of the LM universe of words, yet they account for more than 33% of the negative words which appeared in American 10-Ks (Loughran & McDonald, 2011).

Based on their analysis, Loughran and McDonald (2011) argued that the readability of financial documents could be a reliable predictor of return volatility, as well as a predictor of forecast errors and earnings forecast dispersion among financial analysts. But the biggest takeaway from the LM dictionary is that language matters, especially in times of stress or turmoil, and that many organizations have a problem with communicating in a way that is transparent, straightforward, and open.

The LM dictionary is most pertinent to investor relations and the communication of financial results. But these types of words creep into all types of organizational communication, and not just publicly traded companies. There are important lessons here for internal communications as well, which we will explore more closely in the next section.

Text Mining and Sentiment Analysis

The LM dictionary project is part of an emerging field known as text mining, sometimes referred to as natural language processing. Text mining is a research technique that uses computer algorithms to extract useful information and patterns from large amounts of textual data (Das et al., 2019). Sentiment analysis, like that done by Loughran and McDonald, is a sub-field of text mining, which focuses on the sentiments or opinions contained in a piece of text (Liu, 2020). Some researchers are using text mining and sentiment analysis to study the communications patterns of organizations to see if, among other things, they can create a possible early warning system of fraudulent business activity (Das et al., 2019).

The Enron debacle provides a riveting example of what this research can yield. In the aftermath of Enron’s implosion, the US Federal Energy Regulatory Commission released a treasure trove of information, including Enron’s corporate communications and internal emails for a two-year period leading up to the firm’s collapse. In an analysis of over 100,000 emails sent by Enron employees, as well as over 1000 articles that appeared on PR Newswire from January 2000 to December 2001, Das et al. (2019) found that these corporate communications effectively predicted the crisis. They did this by measuring how positive sentiment declined both internally and externally. But even more interestingly, they found that certain structural characteristics, such as average email length and number of emails sent, were even stronger predictors of trouble than the sentiment analysis. A striking finding was, for one 13-week period, that for every 20-character decline in email length, there was a 1.2% drop in stock price – in fact, email length declined by 50% into 2001. However, even though emails were shorter, senior executives at Enron communicated more frequently, a trend presaging the coming collapse (Das et al., 2019).

The beauty of textual analysis is that it avoids the privacy concerns around reading individual employee emails because its treats datasets systematically and searches for broader trends. As such, Das et al. (2019) believe that regular sentiment and structural email analysis would be an important risk management strategy for organizations and regulators.

The Importance of Clarity

My purpose with this chapter is not to condemn certain words or to urge communicators to police their language more rigidly. Rather, I want to encourage communicators become more curious about the words that they use and what they say about our organizations. When weak modal words come up in our communication, for example, we need to get curious about why we are using them, what kind of information we might be seeking to avoid acknowledging, and whether we – and our organizations – would be better served by choosing words that are more straightforward and candid. After all, we can see how sentiment analysis provides important clues as to what is going on in an organization and text mining proves that organizational communication is a uniquely effective barometer of an organization’s health.

But curiosity also allows us to laugh. Some of the euphemisms, verbal gymnastics, and linguistic contortions that politicians and companies come up with are really very funny, even as they can be infuriating – particularly when they signal deep divisions and outright dishonesty. This is one of the reasons that former Financial Times columnist Lucy Kellaway’s scathing takedowns of corporate language, or what she referred to as claptrap or guff, were so satisfying.

From a neuroscience and psychology perspective, what can we learn from this? What might be going on in the brain and body when we hear weak or strong modal words, or receive emails that are curt and inscrutable? For me, the first thing that comes to mind is ethos, in the Aristotelian framework. The integrity of the communicator is called into question. This ties to the structural dynamics language of Power, which, when ascendant is about competence, optimism and skill, but its shadow can be “win-at-all-costs” and corrupt (Kantor, 2012).

It also says something about how we deal, as individuals and organizations, with vulnerability. There is a type of vulnerability, described so well in Brené Brown’s (2019) work, which is emotionally authentic, facilitates connection, and enables compassion. But there is a less courageous type of vulnerability, that uses evasion and even deceit to conceal itself. This type of vulnerability, even when it’s subtle, triggers a powerful threat response in the brain, causing that sense of something not being quite right. It threatens our need for Certainty, but it can also tread on our needs for Autonomy and Fairness.

Organizations, like people, often have defense mechanisms that protect them from facing uncomfortable or difficult emotions (Obholzer & Roberts, 2019). But just like for people, true well-being comes from facing our demons, acknowledging our fears, and summoning the courage to meet the challenges we encounter. This is the place where transformative growth comes from and is the topic of our next chapter.

- Dr. Laura McHale (PsyD, CPsychol, ABC) is a leadership psychologist, lecturer, and author specializing in neuroscience, communication, and organizational culture. She is the founder and Managing Director of Conduit Consultants, a leadership consulting firm based in Hong Kong. Prior to becoming a psychologist, Dr. McHale worked for over 15 years as a corporate communications executive in the financial services industry, across the US, Europe, and Asia. She now spends her time helping leaders and organizations improve communication, increase performance, boost engagement, and build greater stress resilience.

- Find out more about the book, and buy it on Amazon.


Aabo, T., Als, M., Thomsen, L., & Wulff, J. N. (2020). Watch me go big: CEO narcissism and corporate acquisitions. Review of Behavioral Finance, 43 pages. Akstinaite, V., Robinson, G., & Sadler-Smith, E. (2019). Linguistic markers of CEO hubris. Journal of Business Ethics, 167 (4), 687–705.

Bramley, N. R. (2001). Pronouns of politics: The use of pronouns in the construction of ‘self’ and ‘other’ in political interviews. Ph.D. thesis. Australia National University.

Brown, B. (2019). Dare to lead: Brave work, tough conversations, whole hearts. Random House.

Brozin, M. (2010, September). The intentions behind Barack Obama’s strategic use of personal pronouns. C-thesis English linguistics.

Bull, P., & Fetzer, A. (2006). Who are we and who are you? The strategic use of forms of address in political interviews. Text Talk, 26(1), 3–37.

Das, S. R., Kim, S., & Kothari, B. (2019). Zero-revelation RegTech: Detecting risk through linguistic analysis of corporate emails and news. The Journal ofFinancial Data Science, 1(2), 8–34.

Filimonova, E. (Ed.). (2005). Clusivity: Typology and case studies of inclusive- exclusive distinction. John Benjamin.

Greene, J. (2008). Hostis humani generis. Critical Inquiry, 34 (4), 683–705. Ham, C., Seybert, N., & Wang, S. (2018). Narcissism is a bad sign: CEOsignature size, investment, and performance. Review of Accounting Studies, 23, 234–264.

Heifetz, R. A., Grashow, A., & Linsky, M. (2009). The practice of adaptive lead- ership: Tools and tactics for changing your organization and the world. Harvard Business Press.

Kantor, D. (2012). Reading the room: Group dynamics for coaches and leaders. Jossey-Bass.

Karlsson, S. (2017). Passing on the torch: Discourse strategies in the inaugural speeches of Jiang, Hu and Xi. Master’s thesis, Uppsala University.

Liu, S. (2020) Document-level sentiment analysis of email data. Ph.D. thesis, James Cook University.

Loughran, T., & McDonald, B. (2011). When is a liability not a liability? Textual analysis, dictionaries, and 10-Ks. Journal of Finance, 66(1), 35–65.

Loughran, T., & McDonald, B. (2016). Textual analysis in accounting and finance: A survey. Journal of Accounting Research, 54, 1187–1230.

Obholzer, A., & Roberts, V. Z. (2019). The unconscious at work: A Tavistock approach to making sense of organizational life. Routledge.

Pennycook, A. (1994, April). The politics of pronouns. ELT Journal, 48(2), 173–178.

Xi, J. (2021, January 25). Special address by Chinese President Xi Jinping at the World Economic Forum [Speech transcript]. Via video link from Beijing, China. Retrieved August 26, 2021, from lish/2021-01/25/c_139696610.htm

Warnick, Q. (2010). A close textual analysis of corporate layoff memos. Business Communication Quarterly, 73, 322–326.

BPS Members can discuss this article

Already a member? Or Create an account

Not a member? Find out about becoming a member or subscriber