Floods, Waves, and Surges: The Representation of Latin@ Immigrant Children in the United States Mainstream Media

by Megan Strom et al.

Audio version created with Paper2Audio.

Listen on Paper2Audio

Megan Strom et al.

Audio by Paper2Audio.

Floods, waves, and surges: the representation of Latin@ immigrant children in the United States mainstream media

Abstract

During the 2014 fiscal year, the United States saw a dramatic increase in Latin@ child immigration in hopes of parent-child reunification. The United States mainstream media reacted by reporting heavily on the child arrivals during the summer of 2014. The current study follows a Critical Discourse Studies approach to unveil the ideologies communicated through this media coverage by analyzing the lexical and grammatical representation of Latin@ immigrant children in two of the most read newspapers nationwide: The New York Times and the Los Angeles Times. The results corroborate many previous studies concerning the representation of immigrants in mainstream media by demonstrating how the use of a water-based lexicon and particular verbal processes of members of the majority group vis-à-vis immigrant children communicates an ideology of dealing with and removing the 'problem', rather than, for example, providing aid to the children.

Introduction

Although immigration from Latin America to the United States had leveled off after the financial crisis of 2007 to 2009, starting in Fall 2013, there was a sudden increase in Latin@ children immigrating unaccompanied to the United States from Central America. Although there is no way to know exactly how many of these children have made the harrowing journey alone from their homes in El Salvador, Guatemala, and Honduras, estimates place the total around 90,000 as of Fall 2014. Given that the mainstream media play a key role in determining the way in which a nation reacts to the arrival of at-risk immigrants, it is inherent to understanding how the United States English-language media have represented Latin@ immigrant children. One telling example of how the mainstream media throughout the world represent at-risk groups is the way that the Malaysian press represented refugees and asylum seekers. Mohd Don and Lee (2014, pages 701 to 702) demonstrated how the mainstream media chose two representations for this group:'a major problem that has to be dealt with, and victims that need help – each representation reflecting different interests and fulfilling different agendas'. This is just one example of many in the mainstream media throughout the world that illustrates the disparate representations of immigrants, particularly those who are escaping turmoil in their home countries. In this situation, it is reasonable to expect that many readers of the Malaysian press would adopt one, or both, of these representations of refugees and asylum seekers (although audiences are free to choose any interpretation of social actors they wish). In this way, it is not difficult to reason that readers would perceive immigrants as a problem that needs to be dealt with or as a situation with humanitarian implications, respectively.

In the context of the current study, the question that we address is how the mainstream media in the United States represent unaccompanied Latin@ children. Do they represent these children such that audiences view them compassionately; do they represent Latin@ child immigrants as a problem to be dealt with; or do the media represent them in an entirely different way?

While the representation of immigrants in the mainstream media has been a central focus of Critical Discourse Studies, few studies have considered the way that mainstream media represent children immigrants. This observation points to two large gaps in the literature, particularly in the context of the United States. Without this knowledge, we cannot work toward the central tenet of Critical Discourse Studies: social justice through the identification of social problems and the understanding of how they are represented discursively.

In the current study, we follow a Critical Discourse Studies approach to analyze the representation of Latin@ child migrants in two of the largest newspapers in the United States: The New York Times and the Los Angeles Times. Our results will serve as an initial understanding of whether the United States mainstream media represent an at-risk group as a problem that needs to be addressed, as victims that deserve humane treatment, or if they are represented in an entirely different way. With these data, we will be able to determine what changes, if any, The New York Times and the Los Angeles Times should make to ensure that their representation of Latin@ immigrants, particularly child immigrants, does not perpetuate the negative representation of immigrants that occurs in the United States mainstream media.

CONTACT Megan Strom ✉️ strome02@luther.edu © 2017 Informa UK Limited, trading as Taylor & Francis Group

Background

To understand the phenomenon of Latin@ child migrants arriving in the United States, we must examine the immigration patterns of these children's parents. Historically, single Mexican men who were employed in agriculture constituted the largest group of Latin@ immigrants arriving in the United States under the mandates of the Bracero program. Between 1942 and 1964, this program aimed to keep the number of Latin@ women and children to a minimum to reduce government costs to care for them. After the termination of the Bracero program, more women and children began to arrive, particularly as a result of political turmoil, civil war, and gang violence in Central America (Gutiérrez, n.d.). As a result,'Between 1970 and 1990 the population of migrants from Mexico and Central America grew dramatically from 873,624 to 5,431,992'. By 2000, the number of foreign-born Latin@s living in the United States had reached 14.1 million; just seven years later, that number had increased to 18 million.

These statistics serve as a backdrop for the arrival of thousands of Latin@ child immigrants between the end of 2013 and the summer of 2014 (Krogstad & Gonzalez-Barrera, 2014). These children were fleeing extreme violence and poor economies and were spurred by rumors throughout Central America that Border Patrol would not deport them upon arrival in the United States. Seventy-five percent of these child immigrants came from Central America, specifically El Salvador, Guatemala, and Honduras. Border Patrol detained 47,107 children from October 2013 to May 2014, a figure that was twice as high as the previous fiscal year. However, it is impossible to know exactly how many children came to the United States during this time as these numbers reflected the children who Border Patrol apprehended, and excluded the children in the Border Patrol did not stop. Estimates from October 2013 to October 2014 were around 90,000 children crossing the U.S.-Mexico border without parents or guardians.

Theoretical background

Immigrants in the mainstream media

Critical analyses of the discursive representation of social actors have focused largely on the ways in which the majority media represent minoritized groups. These studies have found that the mainstream media frequently present majority groups in a positive light and minoritized groups in a negative light, thereby reinforcing social hierarchies that support the dominant position of the majority group and the subordinate position of minoritized groups.

Coverage of minoritized groups in the United States mainstream media tends to spike during the following (often related) situations: the arrival of large groups of immigrants, economic recessions, and immigration-related legislation that affects the larger population. For example, through an analysis of the covers of 10 popular magazines published in the United States from 1965 to 1999, Chavez (2001) has found that these media increased coverage of immigrants and immigration during economic crises and when legislation was passed that intended to place more restrictions on immigration. The author notes that the media provided limited coverage of immigration from 1965 until the strong economic recession of the early 1980s, during which magazine covers represented minoritized groups, particularly Mexicans, in a highly negative fashion. Given that much of the coverage regarding minoritized groups played on the fears of Anglo Americans that immigrants would steal jobs and compromise the American way of life during an economic downturn, it is no surprise that the magazine covers Chavez (2001, p. 131) analyzed increased their negative representation of minoritized groups following the 4 November 1986, signing of the Immigration Reform and Control Act by President Reagan. Chavez (2001, p. 193) notes a final increase in the coverage of minoritized groups on magazine covers in the mid-1990s, again corresponding to the passage of anti-immigration legislation: The Illegal Immigration Reform and Immigrant Responsibility Act of 1996.

In short, these magazine covers reflected the economic situation of the country and the legislative work carried out by the government in terms of the harm that immigrants do to the country and the desire for the government to control this harm. Regarding the current study, it will be necessary to keep in mind that the increase in child immigration follows a period of extreme economic stagnation, therefore negative representations of immigrants would not be surprising.

More recently, the United States mainstream media have increased its coverage of immigration following the passage of Arizona Senate Bill 1070 (S.B 1070). This coverage is not surprising given that it followed the most severe economic crisis in the United States since the Great Depression and involved legislation that affected the population at large. Although federal immigration law in the United States trumps state legislation, Arizona passed a bill in April 2010 that appeared to circumlocute federal law by requiring local law enforcement to carry out steps prohibited by the U.S. Constitution. Surprisingly, in a study of the coverage by The New York Times of S.B 1070, Chavez and Hoewe (2012) find that the newspaper contrasted public opinion by presenting distaste toward Arizona's bill. The authors argue that this discursive emphasis on a pro-immigration opinion could potentially have an impact on the long-term public opinion of Senate Bill 1070. Chavez and Hoewe (2012) also find that The New York Times sympathizes with immigrant stories, which they attribute to the demonstration of political orientation of the newspaper or a future change in public sentiment.

Based on this literature, it would not be surprising to see increased coverage of Latin@ child immigrants for two of the aforementioned reasons: it constitutes a major immigration event in which thousands of immigrants arrived in a relatively short period of time, and it occurred on the heels of a disastrous financial crisis. What remains to be seen are the discursive characteristics of this coverage, as this can paint a clearer picture of how the print media represent immigrants.

Immigrant children in the mainstream media

Previous studies concerning the representation of immigrants in the media have focused exclusively on adult immigrants, with the exception of Teo (2000) and González and Torres (2011), the latter only addressing this topic peripherally.

Teo (2000) studies the representation of Vietnamese youth immigrants who form part of a gang called 5 T in Australian print media. His data consist of nine articles published in 1995 in the two newspapers with the largest circulation in Sydney: The Sydney Morning Herald and The Daily Telegraph. Teo follows a critical discourse analytic framework to analyze newspaper headlines and leads, generalization, quotation patterns, over-lexicalization, transitivity, thematization, and lexical cohesion. He finds that the data support ideas emphasized in Critical Discourse Studies regarding the representation of minoritized groups: several discursive constructs essentialize all Asian immigrants in Australia to a single, stereotyped group of highly dangerous gang members. This representation reinforces the positive Us (white Australians) representation and negative Them (Asian immigrants) representation.

González and Torres (2011) report a similar case that reinforces the notion that discriminatory representations of minoritized groups, including children, are not a contemporary phenomenon. The authors narrate the history of young Latin@ Zoot Suiters in Los Angeles who were represented in a discriminatory and racist manner in the mainstream press in the early 1940s. The death of José Díaz in Los Angeles in August 1942 led the media into a frenzy, reducing all Latin@ youth to dangerous gang members, much as the Australian press did in Teo's (2000) study. González and Torres (2011, p. 272) note that 'hostility toward Mexicans was so ingrained in the public's mind' that by the time white marines and sailors attacked the Zoot Suiters in Los Angeles in early June 1943, the media quickly turned the story around to blame the Latin@s with headlines like '44 Zooters Jailed in Attacks on Sailors'.

Initially, then, the limited number of critical discursive analyses of the representation of minoritized youth points to the need of more work in this area for several reasons. First, as Santa Ana (2012) notes, immigration is part of the new globalized economy, and as such, will undoubtedly continue to occur in the near future. Second, the work of Teo (2000) and González and Torres (2011) demonstrate that the media do not appear to discriminate between child immigrants who have committed crimes and those who have not. These findings suggest that the media's'problem' with immigrants is their presence in the host country, a notion that we address in the following section that outlines studies focusing on metaphors that demonize immigrants not for crimes they have committed, but rather for simply entering the host country in large numbers.

Immigrants as a flood

The previous sections have shown that, in general terms, the mainstream media represent immigrants – adult and child – in a negative fashion. Frequently, the media use a highly productive discursive structure that facilitates the cognitive jump from a tangible, concrete notion to a more abstract notion: metaphor. This section focuses on one such of these metaphors that equates immigrants with a flood. However, we first begin by explaining how metaphor can inform our understanding of the discursive representation of ideology.

In the simplest terms, metaphors allow us to understand one aspect of a concept in terms of another (Lakoff & Johnson, 1980). Santa Ana (2002) argues that metaphor is a cognitive framework for understanding the world; it shapes how we understand even the most mundane actions in which we partake on a daily basis. Metaphors constitute a central place in discourse, as they are the tool with which we are able to understand abstract and complex ideas in more simplistic terms. Ideologies, or discursively expressed beliefs held by groups, are an example of the complex ideas that are represented discursively through metaphor. Charteris-Black (2006, p. 573) gives the example of a complex notion, or ideology, that, without the help of metaphor, would likely be difficult to understand in everyday discourse: the fear of the lack of control of'the rate of historical change in a society over time'.

However, by mapping the abstract concept of controlling social change onto a more concrete one, controlling the movement of people, we are more easily able to process and understand this ideology. This mapping of abstract concepts onto concrete ones is the process for creating metaphors, giving us an easier way to understand ideologies expressed through discourse. The following studies give examples of complex ideologies regarding immigrants, for example that immigrants are dangerous and can wreak havoc on a host country, that are represented through metaphors that use water-based terminology as the concrete element upon which they map for easy discursive consumption.

Metaphors that represent immigrants as floods or other water-based natural disasters have been attested in analyses of U.S, British, Malaysian, and Spanish mainstream media. As Chavez (2001, pages 73 to 74) explains,

Another way to indicate a large volume of immigration is through the use of the metaphor of the 'flood' ... A flood, of course, is a deluge that occurs when too much water flows into an area. A flood is not a normal occurrence, but an event that is a problem by definition and one for which a solution must be found as quickly as possible. Floods often wreak havoc on an area and the people living there. As a metaphor for immigration, the flood, along with 'flow', another water image, has come to stand for the movement of people into the country. But whereas a flow may be of a greater or lesser degree, a flood is conceived of as large in magnitude and uncontrollable.

Through this description, Chavez (2001) underscores that associating immigrants with a flood establishes the negativity of immigration because, like floods, it is abnormal and must be stopped. Santa Ana (2002) and van Dijk (2000) further argue that this negative association dehumanizes immigrants by reducing them to climatological phenomena, and thus results in the justification of their negative treatment. Ultimately, the mainstream media throughout the world call upon the fear of the readers of natural disasters to create fear of immigrants and the desire for the government to intervene before the 'flood' wreaks havoc on the general population.

Two studies have considered the use of the flood metaphor in the representation of Latin@ immigrants in U.S mainstream news reporting. Santa Ana's (2002) study closely aligns with the present contribution given that he follows a critical discourse analytic framework to examine the ways in which a U.S mainstream media represent Latin@ immigrants. The author analyzes the use of metaphor to represent Latin@ immigrants in newspaper coverage of anti-immigration legislation in the 1990s in California. This analysis yielded three metaphors that shaped the way the American public viewed these immigrants – Immigration as Dangerous Waters, Immigration as Invasion, and Immigration as Disease or Burden – of which the immigration as dangerous waters metaphor was the most frequently occurring in the dataset. The Los Angeles Times used the immigrants as dangerous waters to represent Latin@ immigrants much as other newspapers across the world have: it robbs immigrants of their individuality, transforming them into 'undifferentiated mass quantit ies ', thereby completely dehumanizing them. Given that natural disasters generally cause a sense of emergency and desperation to keep them at bay by the general public, it is not hard to imagine how the immigration as dangerous waters metaphor would facilitate this kind of reaction from the public with regards to immigrants.

Massey and Pren (2012) considered the negative metaphors'crisis','flood', and 'invasion' as they co-occurred with terms used to identify immigrants such as 'undocumented' and 'Mexican' in The New York Times, The Washington Post, The Wall Street Journal, and the Los Angeles Times starting in 1965. They find that, prior to 1965, there was almost no mention of these negative metaphors to describe Mexican immigrants; however, the metaphors appeared more frequently after the passage of the Immigration and Nationality Act and throughout the remainder of the twentieth century. This corroborates Chavez's (2001) data by demonstrating how media increased their representation of minoritized groups when legislation was passed that intended to place more restrictions on immigration while also showing that, discursively, this coverage makes wide use of anti-immigration metaphors, including those that liken Latin@ immigrants to water-related natural disasters.

Methodology

The current study is a critical analysis of the representation of Latin@ immigrant children in two of the newspapers with the highest circulation in the United States, The New York Times and the Los Angeles Times. We chose these newspapers because their widespread readership indicates their potential for influencing the greatest number of readers through the way that they represent unaccompanied Latin@ child immigrants. The New York Times circulates in print and digitally over two million copies daily, while The Los Angeles Times averages 1.4 new digital readers daily.

We selected our data through a multiple step process. Initial searches indicated that these newspapers started covering the immigration of Latin@ child immigrants in 2014. In order to arrive at a dataset that would warrant a close linguistic analysis, we decided to limit the dataset by including only those articles that directly addressed Latin@ child immigration. To do so, we used the National Newspapers Core Database to search the following terms in news articles from The New York Times and the Los Angeles Times published between January 2014 and December 2014: 'unaccompanied minors', 'Latino migrant children', 'unaccompanied youth', 'unaccompanied migrants', and 'youth migrants'. 'Unaccompanied minors' was the most productive term, yielding 1802 hits, while the other terms yielded 974, 907, 550, and 280 hits, respectively.

After examining the results of these searches, we determined that 'unaccompanied minors' led to a dataset whose articles most closely reflected our goals of completing a micro analysis of the lexical and grammatical representation of Latin@ child immigrants in news articles that covered this event. This resulted in 38 articles from The New York Times and 51 articles from the Los Angeles Times. Our corpus consisted of 80,976 words, with 34,106 words from The New York Times and 46,870 words from the Los Angeles Times.

The analytical framework for this study is based on van Dijk's (1998, 2008a) socio-cognitive approach to Critical Discourse Studies. Following van Dijk's approach, we assume a dialectical relationship between discourse and society whereby we can analyze a social construct, for example ideology, through discourse. The social element we analyze, ideology, is defined as a discursively expressed belief held by groups based in cognition. Following this framework, ideologies are (re) enacted through discourse; therefore, we carry out a discursive analysis in order to shed light on the ideologies regarding Latin@ child immigrants published in the United States mainstream media. In particular, we focus on water-based terminology given its prevalence in the dataset, keeping in mind that previous studies have demonstrated how mainstream media throughout the world use metaphors related to water to promote the ideology that immigrants are a problem that the host community must deal with (compare Charteris-Black, 2006; Chavez, 2001; Crespo Fernández, 2013; Mohd Don & Lee, 2014; Rasinger, 2010; Santa Ana, 2002).

The research questions that guide the analysis are: How do The New York Times and the Los Angeles represent Latin@ immigrant children lexically and grammatically? and How do these representations compare with those of previous studies regarding immigrants in the mainstream media? Our choice to analyze both lexical and grammatical discursive constructs speaks to the 'eclectic' methods of Critical Discourse Studies. Rather than choosing which elements to study at the outset of our investigation, we have instead chosen to study those discursive constructs which address our objective of understanding the representation of Latin@ child immigrants. Ultimately, our choice to analyze the lexical and grammatical representation of Latin@ child immigrants allows us to arrive at a broader understanding of the overall discursive message our dataset communicates.

As with any critical discourse analysis that focuses on a limited number of discursive features, this study examines closely just two discursive features of the many available for analysis. Therefore, our results paint a necessarily limited picture of the way that The New York Times and the Los Angeles Times represent Latin@ child immigrants; that is, we can only speak to the ways in which the lexical and grammatical items we analyze represent these children. Of course, to arrive at a complete understanding of this representation, we would need to examine other discursive structures in our dataset.

Our analysis consisted of a lexical and a grammatical section. First, we analyzed the ways in which lexical items associated with water represented Latin@ immigrant children. We focused on the terms 'surge', 'wave', and 'flood', as these terms came to the fore during a first pass through the dataset.

We acknowledge that other metaphors, such as the Immigration as Invasion and Immigration as Disease or Burden metaphor that Santa Ana (2002) attested, are certainly present and should be considered in future analyses in order to arrive at a better understanding of this dataset. Second, we determined which terms appeared most frequently in the same sentences as the aforementioned terms. From this, we analyzed how the terms that collocated with 'surge', 'wave', and 'flood' represented Latin@ immigrant children.

For the grammatical analysis, we focused on verbal processes and the social actors who carried out these processes. First, we identified which participants, or social actors, carried out actions associated with the'surge/wave/flood'. Following Richardson (2007), we then determined the verbal processes that these social actors carried out, categorizing them as mental (for example to think), relational (for example to be), or material (for example to kick). From this grammatical analysis, we were able to determine how the verbal processes in our dataset were indicative of similar or different ideologies observed in the lexical analysis.

Results

Lexical analysis

Water-based vocabulary

Chavez (2001, p. 16) has noted,'The flood is a common metaphor for the flow of immigrants to the United States. And with floods comes danger as the raging waters overwhelm people, land, and nations'. Our data corroborate this idea. In our dataset of 89 articles from The New York Times and the Los Angeles Times,'surge' appeared 91 times,'flood' appeared 21 times, and 'wave' appeared 14 times (see Table 1). This water-based terminology establishes a metaphor that represents immigrants as floods. Consequently, these representations call upon ideologies of immigrants as natural disasters who should be dealt with in an inhumane fashion.

Table 1 summary: The term 'Surge' is used significantly more often than 'Flood' or 'Wave' across both newspapers to describe water-related events. Out of a total of 126 occurrences across both publications, 'Surge' appeared 91 times, while 'Flood' appeared 21 times and 'Wave' appeared 14 times. This pattern is consistent in both The New York Times, where 'Surge' appeared 43 times, and the Los Angeles Times, where it appeared 48 times.

The news articles in our dataset used the term 'surge' most frequently to describe the increase in Latin@ child immigration to the United States. The Merriam-Webster dictionary defines a surge as 'a sudden, large increase; a sudden movement of many people; a large wave of water'. While this definition does not inherently point to the negativity of 'surge', the way it is used in context demonstrates how this term equates the arrival of Latin@ immigrant children in the United States with an uncontrollable climatological phenomenon:

(1) Administration officials have been trying to cope with a surge of unaccompanied children that has overwhelmed border officials as well the nation's family and immigration court systems. (The New York Times, 7 June 2014)

(2) The presence of unaccompanied migrant children is not new, but the surge in recent months has overloaded Border Patrol stations and detention facilities, particularly in Texas. (Los Angeles Times, 8 June 2014)

These examples describe the increase in Latin@ child immigration as of Fall 2013 and its implications at the U.S.-Mexico border, particularly for border patrol officials. Rather than describing these children in terms that underscore the violence they suffer and the assistance they need upon arriving in the United States, 'surge' represents them as a natural disaster that cannot be controlled and as one that causes extensive damage. If metaphors shape how we understand the world, it stands to reason that readers will understand Latin@ child immigrants as a 'surge' that is powerful enough to 'overwhelm' border officials, detention facilities, and immigration court.

There are two possible ideological effects of this metaphor. The first is proactive, by communicating an ideology that the United States should attempt to stem the 'surge', just as it would with a natural disaster. Another possible ideological effect is defensive, wherein the audience reacts with fear based on the inability to defend itself, a reaction that is traditionally associated with natural disasters.

This seems particularly viable in light of examples (1) and (2) in which the organizations typically tasked with apprehending and deporting these children find themselves unable to do so. Examples (3) and (4) that follow also lend credence to the latter ideological effect given that they underscore the size and intensity of this 'surge':

(3) In an unprecedented surge, more than 57,000 young migrants... have been apprehended. (The New York Times, 20 July 2014)

(4) a surge of at least 57,000 unaccompanied minors (Los Angeles Times, 10 July 2014)

As Crespo Fernández (2013, p. 99) notes, the media use numbers and percentages to give'credibility to the idea of immigration as something chaotic and out of control. which makes the problem appear more serious',. In examples (3) and (4), the 'surge' is 57,000 people strong, thus making credible the idea established in examples (1) and (2) above that Latin@ immigrant children'overwhelm' border patrol, detention facilities, and immigration court. Consequently, the specificity of how many unaccompanied minors are coming to the United States perpetuates the audience's fear that this'surge' is large and out of control. In this way, it seems probable that the ideology communicated by this water-based metaphor and the numbers that specify it is one in which audiences fear Latin@ immigrant children because they are an out-of-control natural disaster.

Another frequently appearing term in our dataset is 'flood' (see Table 1), which The Merriam-Webster dictionary defines as 'a large amount of water covering an area of land that is usually dry; a large amount of things that come or happen at the same time'. This term functions similarly to 'surge' in our dataset in that it likens Latin@ child immigrants to an uncontrollable climatological phenomenon. In example (5), 'flood' establishes a metaphor in which Latin@ immigrant children are a natural disaster:

(5) The president said he needed the money to set up new detention facilities, conduct more aerial surveillance and hire immigration judges and Border Patrol agents to respond to the flood of 52,000 children. (The New York Times, 9 July 2014)

This example was published immediately following the peak of apprehensions of Latin@ child immigrants in 2014. Through its use of a water-based metaphor, this example paints a picture of child immigrants not as a group who has suffered immensely before arriving at the United States southwest border and thus needs immediate aid, but rather as a group that arrived unexpectedly and needs to be dealt with. Moreover, the addition of the strength of the 'flood', that is, '52,000 children', highlights the chaotic nature of this natural disaster. Ideologically, these examples guide audiences to perceive Latin@ immigrant children in the same way that the mainstream media represent adult immigrants. By representing these children as floods, example (5) dehumanizes them and discursively justifies their inhumane treatment. The metaphor in example (6) carries out the same ideological work by equating child immigrants with a flood:

(6) He said the authors of the law did not foresee that children and teens would flood the border and overwhelm emergency shelters and child-care facilities, or that immigration courts would clog so quickly. (Los Angeles Times, 6 July 2014)

Of note here is the use of 'overwhelm' and 'clog', which reinforce the water metaphor by calling upon well-known characteristics of any flood. Through this metaphor, then, these children become forces of nature capable of overwhelming all 'lines of defense', that is, shelters, child-care facilities, and immigration courts. Again, this notion of Latin@ child immigrants as a natural disaster promotes the same sort of reaction the country would have to a flood, which would typically include measures to decrease or expel the flood.

The third most frequently appearing water-related term in our dataset was 'wave' (see Table 1), which functions similarly to 'surge' and 'flood' by portraying Latin@ immigrant children as a natural disaster:

(7) Cristian is one of the tens of thousands of unaccompanied minors who have illegally crossed the border with Mexico in recent months, in a wave that has overwhelmed immigration officials and prompted the Obama administration to declare a humanitarian crisis.

(8) The detainees are part of a wave of Central American immigrants who have overwhelmed Border Patrol facilities in southern Texas after crossing the border illegally.

Examples (7) and (8) represent unaccompanied minors as a 'wave' so strong that it has the power to 'overwhelm' immigration officials and border patrol facilities. Describing the children's crossing as 'illegal' in these examples supports the view that this 'wave' is inherently negative and, as such, undesired. Of note in example (7) is the representation of one immigrant child in particular, 'Cristian', by means of his first name. This is an aspect of our dataset that deserves further investigation in light of the positive associations that occur when social actors are represented as individuals through the use of their name in news reporting.

If metaphors shape the way we perceive the world, it would follow, then, that these examples appearing in The New York Times and the Los Angeles Times establish a world view wherein the arrival of Latin@ immigrant children is negative and must be stemmed. Consequently, it is not likely that readers will insist on humane treatment for these children, but rather their expeditious deportation.

In conclusion, the examples above illustrate how the frequently occurring terms 'surge', 'flood', and 'wave' represent Latin@ immigrant children as a climatological phenomenon that overwhelms even the last line of defense at the United States southwest border. What is more, the inclusion of numbers with these terms underscores the chaotic nature of these water-based disasters. The ideological thrust of these metaphors is negative given that they liken children to a dangerous natural disaster that is to be feared because of its strength and the inability of the nation to impede it, rather than an ideology that emphasizes the desperate need for humane assistance for these children. The next section demonstrates how the lexical items that most frequently co-occur with 'surge', 'flood', and 'wave' reinforce the ideology that Latin@ children are a dangerous natural force to be feared.

Collocation. For Fairclough (2001, p. 95), lexical items that co-occur, or collocate, can lead to an ideologically motivated classifying scheme. That is, the co-occurrence of terms establish a framework for understanding a particular topic. For example, in the current dataset, a set of words that appear together frequently frame Latin@ child immigration as a crisis that overwhelms authorities. This framework communicates an ideology whereby authorities should treat the children as they would a 'surge/flood/wave'.

In the current study,'crisis' co-occurred with the water lexicon 18 times;'emergency' co-occurred with the water lexicon nine times; and'overwhelm (ed)' co-occurred eight times. All three collocations can be found with 'surge' in example (9):

(9) With border authorities in South Texas overwhelmed by a surge of young illegal migrants traveling by themselves, the Department of Homeland Security declared a crisis this week and moved to set up an emergency shelter for the youths.

Similar to examples (1 to 8) above, example (9) underscores how the terms that collocate with the water lexicon establish an ideologically motivated classifying scheme for Latin@ immigrant children. This scheme is based on the notion that the youth are a 'surge', that is, a natural disaster, and therefore can 'overwhelm' border patrol. If border authorities, which most Americans consider the country's last defense against immigrants, are 'overwhelmed', the situation must be alarming enough that it warrants calling it a 'crisis' that requires 'emergency' response.

Similar to examples (7) and (8) above, example (9) also utilizes 'illegal' to classify child immigrants and to highlight their inherent negative characteristics. Thus, this example discursively represents unaccompanied Latin@ migrant children who are fleeing dire situations in their home countries as a natural disaster for the United States. By representing border patrol as 'overwhelmed' by this 'surge', the children's need for help is discursively shifted to border authorities. In this way, these examples lexically reinforce the notion that Latin@ child immigrants are natural disasters who should be treated as one would treat a 'surge/wave/flood'.

To conclude the lexical analysis, in this dataset, the terms 'surge', 'flood', and 'wave' establish a metaphor that represents Latin@ youth migrants as a natural disaster. The addition of exact numbers of immigrants demonstrates the intensity of the 'surge/flood/wave' and reinforce the notion that this natural disaster was out of control. The terms that occurred most frequently with the water lexicon, 'crisis', 'emergency', and 'overwhelm(ed)', establish a framework through which readers can understand Latin@ child immigration. This framework reinforces the ideology that arises from the water lexicon by underscoring the helplessness of border authorities in the face of this crisis.

Thus, the words in this dataset pattern around a negative representation of Latin@ child immigrants. Ideologically, it is likely that this lexical pattern communicates fear of youth immigrants, particularly in light of the emphasis on the inability of United States border patrol to prevent or stem their arrival.

It bears repeating that the results of this lexical analysis only represent a part of our dataset, namely those sentences that contain the terms 'surge', 'wave', and 'flood', as well as the sentences containing collocating terms. Thus, our conclusions are limited solely to these sections of the dataset. In order to arrive at a better understanding of how these newspapers represent Latin@ child immigrants, more discursive elements must be analyzed. We present a grammatical analysis in the next section to shed more light on the ideologies communicated by 'surge/wave/flood'.

Grammatical analysis

While the previous section focused on the ideologies that arose from the lexical representation of Latin@ immigrant children, the present section focuses on the ideologies that arise from the grammatical representation of these children. Given that different discursive constructs communicate different ideologies, the objective of this section is to offer another perspective regarding the representation of Latin@ child immigrants in our dataset. In particular, we analyze grammatical agency because it is indicative of ideology.

For this analysis, we identified which participants, or social actors, carried out actions on the 'surge/wave/flood' of Latin@ immigrant children. Following Richardson (2007), we then categorized these actions as mental (for example to think), relational (for example to be), or material (for example to kick). From this grammatical analysis, we were able to compare the ideologies that arose from the lexical analysis to the ideologies represented grammatically in our dataset.

Two groups of participants emerged from this analysis: Latin@ children immigrants and those opposing them. Initially, it seemed that the presence of these groups established a social binary similar to what van Dijk (1998, 2008b) has noted for mainstream media throughout the world consisting of the majority group (for example Anglo Americans) and a minoritized group (for example Latin@ children immigrants). More specifically, the most frequently occurring social actors for the majority group represented authority figures ('President Obama', 'Congress', 'Border Patrol'), while the most frequently occurring social actors for the minoritized group emphasized that these children were alone ('unaccompanied') and young ('minors', 'children'). We now explore how verbal processes carried out by the majority group reinforce the lexically established ideology wherein Latin@ child immigrants are inherently negative and must be returned to their home countries.

One of the participants who most frequently carried out actions associated with the'surge/wave/flood' was border patrol. The following is an official description of this federal law enforcement agency: its primary mission is'to detect and prevent the illegal entry of aliens into the United States... Undaunted by scorching desert heat or freezing northern winters, they work tirelessly as vigilant protectors of our Nation's borders', (U.S. Customs and Border Protection, n.d.). From this description, it appears that no weather event can keep these officers from carrying out their job of protecting the United States borders. However, as examples (10) and (11) demonstrate, the verbal processes associated with border officials make it clear that they are unable to confront the'surge/wave/flood' of Latin@ child immigrants arriving at the United States southwest border:

(10) frontline Border Patrol stations there were often overwhelmed

(11) Border officials are bracing for another wave of children as the weather cools.

Here border patrol are agents of the relational process 'to be (overwhelmed)' and the intransitive action 'to brace' oneself. In fact, in all of the sentences we analyzed for agency, border authorities never appeared as agents of verbs that acted directly upon Latin@ child immigrants. Thus, border patrol is not grammatically powerful because it was never the agent of transitive verbs vis-à-vis the 'surge/wave/flood'.

These types of processes underscore the notion that the natural disaster at the border is so dire that border authorities can do nothing but 'brace' themselves and be 'overwhelmed' by the oncoming natural disaster. In this way, the grammatical analysis corroborates the results of the lexical analysis in that unaccompanied Latin@ children are represented as a phenomenon that renders border officials powerless.

Other government officials located in Washington, D.C. were the participants of verbal processes related to the 'surge/wave/flood'. Unlike border patrol, government officials were the subjects of transitive verbs; however, these verbs represented mental processes, rather than processes carried out on other social actors:

(12) a border crisis that both Democrats and Republicans say the White House was slow to notice.

The majority of the verbal processes carried out by government authorities in our dataset were mental processes like the verb'to notice' in example (12). Given that government authorities are simply'noticing' the 'crisis', rather than acting upon it, points to the lack of influence they have over the natural disaster. Thus, grammatical agency calls upon the same notion established through the lexicon wherein powerful groups were unable to stem the arrival of the'surge/wave/flood' of Latin@ immigrant children to the United States.

The only social actor in the data who was frequently the subject of material processes was the most powerful person in the United States, President Barack Obama. The dataset represented President Obama as the only member of the powerful social group who was able to carry out actions against the 'surge/wave/flood' of Latin@ child immigrants:

(13) President Obama on Monday ordered federal emergency authorities to take charge of the relief effort.

(14) President Obama is delivering a message to Central American parents... Don't risk it.

In these examples, President Obama is the subject of the transitive verb'ordered' and of the negative command'don't risk it'. In this way, the president discursively confronts and attempts to stop the'surge/flood/wave' that the other social actors in the dataset were unable to (see examples (10 to 12) above). These examples reinforce grammatically what we concluded for the lexicon because President Obama, the most powerful person in the United States, is the only participant who is represented as fighting to stop the'surge/flood/wave' of Latin@ child immigrants. That is, verbal processes underscore the ideology that child migrants are a dangerous natural disaster that only the most powerful in the nation can address.

To summarize, the grammatical analysis points to unexpected results with regards to previous research in Critical Discourse Studies, while at the same time reinforcing the ideologies that arose from lexical analysis. Initially, the presence of social actors from opposing groups appeared to reify van Dijk's (1998, 2008b) finding that mainstream media throughout the world pit minoritized groups against powerful majority groups. While our grammatical analysis demonstrated that these groups are pitted against one another, the lack of discursive power of the majority group vis-à-vis the minoritized group was unexpected.

This is because border patrol was presented as most reactive to the arrival of child migrants through relational and intransitive verb processes. Similarly, while government officials were the agents of some transitive constructions, these verbs were mental, and did not convey a potential to stem the 'surge/wave/flood' of Latin@ child immigrants. Only President Obama was represented as expected because he carried out material processes that acted directly upon the 'surge/wave/flood'.

Overall, though, the results of the grammatical analysis corroborate the findings for the lexical analysis by representing Latin@ child immigrants as a negative natural disaster that overwhelms traditionally powerful social actors, and can only be contained by the president of the country.

Conclusion

This section addresses the theoretical and critical implications of the lexical and grammatical representation of Latin@ children immigrants in two of the largest newspapers in the United States, The New York Times and the Los Angeles Times, as well as avenues for future research.

With regards to the first research question, through a critical lexical analysis of 89 articles, we have demonstrated how the water-based terms 'surge', 'flood', and 'wave' represented Latin@ immigrant children as a natural disaster. In this way, these children became a chaotic, overwhelming, and uncontrollable force that must be stopped. Therefore, our dataset communicated the ideology that at-risk children are a threat to the United States while authorizing their inhumane treatment. The use of numbers augmented the severity of the natural disaster and reinforced the notion that the arrival of immigrant children should be viewed with the same fear one would view a large surge/wave/flood.

Three terms that co-occurred with the water lexicon, 'crisis', 'emergency', and 'overwhelm(ed)', underscored the chaos associated with natural disasters. These collocating terms communicated that the strength of the 'surge/flood/wave' was so overwhelming that nothing could be done to stop it; only reactionary emergency measures could be put in place to 'deal with' the aftermath.

This ideology was reinforced grammatically given that border patrol and government officials were the subjects of the relational and mental processes, underscoring how these social actors were unable to stem the 'surge/flood/wave' of Latin@ child immigrants. The only social actor who could stand up to the natural disaster was the most powerful entity in the United States, President Obama. Thus, the verbal processes in our dataset perpetuated the ideology that child immigrants should be feared and dealt with as a problem rather than as a humanitarian crisis that requires immediate aid for at-risk children.

Regarding the second research question, our results are similar to previous studies regarding the representation of immigrants in the mainstream media throughout the world, particularly with regard to the association of immigration with a natural disaster that leads to their dehumanization and an acceptance of their negative treatment. This also corroborates the few previous studies regarding the representation of immigrant children in that the mainstream media represent them as a problem that must be dealt with, rather than an at-risk group that requires humanitarian aid. In this way, our study lends strength to the previous studies that have argued that the mainstream media discursively perpetuate the hegemonic structure in which the majority group is represented humanly and positively, and minoritized groups are represented negatively, and as less than human.

However, what previous research has not revealed is how the United States mainstream media represent Latin@ child immigrants. The current study fills this gap by providing data regarding the ways in which two English-language newspapers lexically and grammatically represent a minoritized group. There are several critical implications of a study of this kind.

First, we believe it is important to take into account these findings and continue this line of investigation because similar types of immigration are currently occurring throughout the world, for example, from Africa to Europe and from Central America to Mexico. Our study can provide insight about what the representation of child immigrants might look like in other geographical contexts. Based on what we have found, we would not be surprised to see highly negative representations of youth African immigrants in the European mainstream media and of youth Central American immigrants in the Mexican mainstream press.

Secondly, our study has critical implications for Latin@ immigrants, both adult and child, in the United States. Given the strong role the media have in shaping the nation's ideologies, we would not be surprised if these newspapers led their millions of readers to believe that Latin@ immigrants, even very young child migrants, are dangerous and should be immediately removed from the United States. The use of a lexicon based on water-related natural disaster and the presence of mental and relational verbal processes precludes the need to understand the reasons behind child migration, and simply instructs readers to fear these children. We argue that by focusing on the putative negative characteristics of child migrants, the examples we analyzed from The New York Times and the Los Angeles Times make it convenient for readers to dismiss the plight of these children, making it highly unlikely that readers will support legislation to assist them in their transition to a new life.

Finally, because we have only studied two newspapers, we have no way of knowing how the United States mass media function overall with regards to this topic. However, understanding that much of the mainstream media function under the premise of perpetuating the hegemonic structure, we estimate that many other television outlets are broadcasting similar ideologies. That is to say, millions of American consumers of mainstream media have likely read or heard similar notions on the many platforms available for news consumption today. As such, we take this opportunity to insist that the mainstream media in the United States stop using water-based metaphors to communicate negative ideologies regarding Latin@ immigrants. We have shown in this study that likening children who are fleeing gang violence, torture, abuse, kidnapping, and a multitude of difficulties in their home countries to a dangerous natural disaster perpetuates the already existing negative ideologies regarding immigrants in the United States.

Given that this is on the first studies to address the representation of Latin@ children immigrants in the United States mainstream media, future research will be needed in several areas. First, it will be necessary to analyze other major print publications such as The Wall Street Journal and The Washington Post to determine if the findings from the current study hold across other publications with large readerships in the United States. Second, because ideology is expressed through a multitude of discursive structures, future studies should consider other lexical and grammatical items, context, and how production and consumption factors such as authors, newspaper ownership, and readership pattern with the ideologies represented in the newspapers. In this way, we will arrive at a better understanding of how media that reach millions of consumers on a daily basis portray Latin@ child immigrants. Most importantly, this information will help us determine what the mainstream media should do to represent Latin@s, particularly those children who have fled dangerous situations in their home countries to be reunited with the parents and family in the United States, as a group that should be treated humanely, and not as a natural disaster to be dealt with.

Notes

1. Latin@ youths who wore clothing called Zoot Suits were the targets of racial attacks by Anglo Americans in Los Angeles in the early 1940s.

2. The audit bureau did not offer statistics regarding the daily print circulation for the Los Angeles Times.

Notes on contributors

References

Alliance for Audited Media. (2014). Audience snapshot database. Audience summary database. Retrieved from http://online.audiencefax.com/CreateReport.aspx

Charteris-Black, J. (2006). Britain as a container: Immigration metaphors in the 2005 election campaign. Discourse & Society, 17(5), 563–581.

Chavez, L. (2001). Covering immigration. Popular images and the politics of the nation. Berkeley: University of California Press.

Chavez, M., & Hoewe, J. (2012). National perspectives on state turmoil. Characteristics of elite U.S. newspaper coverage of Arizona SB 1070. In O. Santa Ana & C. González de Bustamante (Eds.), Arizona firestorm. Global immigration realities, national media, and provincial politics (pp. 189–202). New York, NY: Rowman & Littlefield Publishers.

Chin, G. J., Hessick, B., & Miller, M. (2012). Arizona Senate Bill 1070: Politics through immigration law. In O. Santa Ana, & C. González de Bustamante (Eds.), Arizona firestorm. Global immigration realities, national media, and provincial politics (pp. 73–96). New York, NY: Rowman & Littlefield Publishers.

Crespo Fernández, E. (2013). The treatment of immigrants in the current Spanish and British right-wing press: A cross-linguistic study. In M. Martínez-Liola (Ed.), Discourses on immigration in times of economic crisis: A critical perspective (pp. 86–112). Newcastle upon Tyne, UK: Cambridge Scholars Publishing.

Crespo Fernández, E., & Martínez Lirola, M. (2012). Lexical and visual choices in the representation of immigration in the Spanish press. Spanish in Context, 9(1), 27–57.

van Dijk, T. (1988a). How ‘they’ hit the headlines: Ethnic minorities in the press. In G. Smitherman Donaldson & T. van Dijk (Eds.), Discourse and discrimination (pp. 221–262). Detroit, MI: Wayne State University Press.

van Dijk, T. (1988b). News analysis. Case studies of international and national news in the press. Hillsdale, NJ: Lawrence Erlbaum.

van Dijk, T. (1989). Mediating racism. The role of the media in the reproduction of racism. In T. Wodak (Ed.), Language, power and ideology: Studies in political discourse (pp. 199–226). Amsterdam: John Benjamins.

van Dijk, T. (1991). Racism and the press. London: Routledge.

van Dijk, T. (1998). Ideología. Una aproximación multidisciplinary [Ideology. A multidisciplinary approach]. Barcelona: Gedisa.

van Dijk, T. (2000). Ideologies, racism, discourse: Debates on immigration and ethnic issues. In J. ter Wal & M. Verkuyten (Eds.), Comparative perspectives on racism (pp. 91–116). Surrey: Ashgate.

van Dijk, T. (2008a). Discourse and context. A sociocognitive approach. Cambridge: Cambridge University Press.

van Dijk, T. (2008b). Discourse and power. New York, NY: Palgrave Macmillan.

Fairclough, N. (2001). Language and power (2nd ed). New York, NY: Longman.

Fairclough, N., Graham, P., Lemke, J., & Wodak, R. (2004). Introduction. Critical Discourse Studies, 1(1), 1–7.

Fairclough, N., & Wodak, R. (1997). Critical discourse analysis. In T. van Dijk (Ed.), Discourse as structure and process: A multidisciplinary introduction (Vol. 2, pp. 258–284). London: Sage.

González, J., & Torres, J. (2011). News for all the people. The epic story of race and the American media. New York, NY: Verso.

Gutiérrez, D. (n.d). An historic overview of Latino immigration and the demographic transformation of the United States. American Latinos and the making of the United States: A theme study. Retrieved from https://www.nps.gov/heritageinitiatives/latino/latinothemestudy/immigration.htm

Krogstad, J. M., & Gonzalez-Barrera, A. (2014). Number of Latino children caught trying to enter U.S. nearly doubles in less than a year. Retrieved from http://www.pewresearch.org/fact-tank/2014/06/10/number-of-latino-children-caught-trying-to-enter-u-s-nearly-doubles-in-less-than-a-year/

Lakoff, G., & Johnson, M. (1980). Metaphors we live by. Chicago, IL: University of Chicago Press.

van Leeuwen, T. (2008). Discourse and practice. New tools for critical discourse analysis. New York, NY: Oxford University Press.

Martín Rojo, L., & van Dijk, T. (1997). "There was a problem, and it was solved!" Legitimating the expulsion of "illegal" immigrants in Spanish parliamentary discourse. Discourse & Society, 8(4), 523–566.

Massey, D. S., & Pren, K. A. (2012). Unintended consequences of US immigration policy: Explaining the post-1965 surge from Latin America. Population and Development Review, 38(1), 1–29.

Mohd Don, Z., & Lee, C. (2014). Representing immigrants as illegals, threats and victims in Malaysia: Elite voices in the media. Discourse & Society, 25(6), 687–705.

Paret, M. (2014). Legality and exploitation: Immigration enforcement and the US migrant labor system. Latino Studies, 12(4), 503–526.

Pew Research Center. (2014). Unauthorized immigrant population trends for states, birth countries and regions. Retrieved from http://www.pewhispanic.org/2014/12/11/unauthorized-trends/

Rasinger, S. M. (2010). Lithuanian migrants send crime rocketing: Representation of 'new' migrants in regional print media. Media, Culture & Society, 32, 1021–1030.

Richardson, J. (2007). Analysing newspapers. An approach from critical discourse analysis. New York: Palgrave Macmillan.

Santa Ana, O. (1999). 'Like an animal I was treated': Anti-immigrant metaphor in US public discourse. Discourse & Society, 10(2), 191–224.

Santa Ana, O. (2002). Brown tide rising. Metaphors of Latinos in contemporary American public discourse. Austin: University of Texas Press.

Santa Ana, O. (2012). Arizona's provincial responses to its global immigration challenges. Introduction to Arizona firestorm. In O. Santa Ana, & C. González de Bustamante (Eds.), Arizona firestorm. Global immigration realities, national media, and provincial politics (pp. 3–18). New York, NY: Rowman & Littlefield Publishers.

Teo, P. (2000). Racism in the news: A critical discourse analysis of news reporting in two Australian newspapers. Discourse & Society, 11(1), 7–49.

The New York Times Company. (2014). The New York Times marks solid circulation gains. Retrieved from http://investors.nytco.com/press/press-releases/press-release-details/2014/The-New-York-Times-Marks-Solid-Circulation-Gains/default.aspx

U.S. Customs and Border Protection. (n.d.). Border Patrol overview. Retrieved from http://www.cbp.gov/border-security/along-us-borders/overview

Wodak, R., & Meyer, M. (2009). Critical discourse analysis: History, agenda, theory and methodology. In R. Wodak, & M. Meyer (Eds.), Methods of critical discourse analysis (pp. 1–33). Los Angeles, CA: Sage.