Lexical Innovation in Ecotourism Discourse: The Case of Eco(-)lodge

,


Introduction
Ecotourism emerged as an opportunity to re-conceive travelling in times of unprecedented environmental breakdown.Included in the larger category of sustainable tourism, it refers to responsible travel to natural areas that tries to minimise impact on the destination's ecosystem, finance conservation projects, support local communities, and raise ecological awareness in its practitioners.Conservation, education, ethics, sustainability, and local benefits are the variables that are most frequently addressed in its definitions (Fennell, 2001).The former objectives may be imposed on a broad range of activities afforded to aspiring ecotourists, often in the form of adventurous experiences in natural environments, such as kayaking, cycling, hiking, safaris or other encounters with wildlife, etc.Despite the critics' concern with its factual sustainability and the persistence of (human) self-interest as opposed to the well-being of ecosystems, 'ecotourism is alive and well' (Fennell, 2022, p. 2), and actually growing (Precedence Research, 2023).
In terms of language, the specialisation of tourism discourse has long been ascertained (Gotti, 2006;Maci, 2020;Nigro, 2006), and studies have also explored the creativity of its lexical constructions (Frank, 2021;Koval et al., 2023) and suasive features (Manca, 2016) to arouse the interest of potential tourists.However, things ought to be reconsidered in the case of ecotourism communication, since its aim is not limited to the promotion of travel per se (or in order to maintain the host country's revenue flow), but it must encourage conscious behaviour in all actors involved, at all stages.Therefore, and since it has been developing its own 'register' (Dann, 1996a(Dann, , 1996b)), and producing more and more labels to describe the alternative experiences it offers (Raţă et al., 2012), ecotourism discourse represents an interesting opportunity to address the ecolinguistic objective of 'explor [ing] the role of language in the life-sustaining interactions among humans, other species, and the physical environment' (Stibbe, 2021, p. 11).Indeed, given that interpretation, seen as the ability to understand and appreciate natural places and subjects, is one of ecotourism's key principles (TIES, 2019), the quality and performance of ecotourism activity is inextricably tied to the quality and performance of its discursive practices (Mühlhäusler, 2020;Mühlhäusler & Peace, 1999).
Therefore, the present paper investigates lexical innovation in ecotourism discourse in order to assess its function in supporting the ecological ends of the activity.It looks at the specific case illustrated by the noun eco(-)lodge (Note 1), created to advertise alternative types of accommodation in ecotourism destinations.The word features consistently in the official names of several structures, and ecolodges in general are seen as developing their own marketplace (Spenceley & Rylance, 2022).The analysis collocates itself within the theoretical framework of ecolexicography, which acknowledges the contribution that lexicographical practice and outputs can make to environmental protection.Section 2 provides a sketch of the former's history, although it devotes more attention to Liu et al.'s (2021) recent theorisation, as it guided the present research.Furthermore, it reviews some essential literature contributing to the field.Section 3 proceeds to explain the methodology used, which relied on quantitative and qualitative approaches, and considered two main sources of data: dictionaries and corpora.Results are presented and discussed in Section 4. First, the meaning of eco(-)lodge is explored according to available definitions.Second, further insight into its context of use is obtained through the analysis of occurrences in the corpora.Findings from the two sources are compared to reveal a gap between them, offering a starting point for reasoning on common expectations about ecolodges.Whereas defined only with reference to their environmental impact, ecotourism texts mainly connote these buildings as luxury facilities found in "protected" natural contexts, generally overlooking sustainability.This leads to some considerations about the progressive semantic bleaching (Mühlhäusler & Peace, 2001) of the combining form eco-as a driver of lexical creativity.To conclude, section 5 summarises the outcomes of the study.It reflects on the viability of updating current lexicographical materials, and advocates for the need to redress the imbalance between economy and environmentalism evidenced in ecotourism discourse.

Theoretical Background
Ecolexicography was first proposed by Sarmento (2000Sarmento ( , 2001Sarmento ( , 2002Sarmento ( , 2005)), who intended to combine lexicographical research with the concerns of ecolinguistics (Fill & Mühlhäusler, 2001;Ha, 2023;Penz & Fill, 2022;Stibbe, 2021).His programme did not consist in the compilation of a specialised dictionary of ecological terms.Rather, he envisioned a project focussed on the identification, description, and categorisation of words according to their ecological or non-ecological role, that is, the favourable/unfavourable ideological stance they convey with respect to the preservation of the planet.Sarmento did not provide a clear criterion for distinguishing between the two classes of expressions, but simply drew up a list of examples to illustrate his point.For instance, web of life, other animals, and in-oneself-with-the-other were marked as having an ecological role; instead, anthropocentrism, crisis of fragmentation, and Nature were seen as non-ecological.
Sarmento set the agenda for the new (sub)discipline, even though factual contributions referring to his theory were not provided in the following years.The notion of ecolexicography was eventually taken up again by Albuquerque (2018aAlbuquerque ( , 2018bAlbuquerque ( , 2019aAlbuquerque ( , 2019b)).First, he discussed it as an instance of pedagogical lexicography, and conceived it as functional to the achievement of 'ecoliteracy' ('ecoletramento') (Albuquerque, 2018a).He postulated for the education towards ecological sensibility of both learners and teachers, starting from metalexicographic activities aimed at evaluating learners' dictionaries used in class.Later on, he recovered and developed Sarmento's original endeavour (Albuquerque, 2019a(Albuquerque, , 2019b)).In fact, Albuquerque argued for the creation of an 'ecodictionary' ('ecovocabulário') which expands entries to address the consequences that using certain words may have on the relationships among individuals and ecosystems, so as to produce entries which foster an ecological worldview.The latter would consist of four main elements: definitions, examples of usage, effects ('efeitos') and results ('resultados').
More recently, Liu et al. (2021) put forward a new paradigm for doing ecolexicography, which is the one inspiring the present study.They distinguished between a micro and macro level of analysis.Micro-ecolexicography sees dictionaries as complex and 'integrated whole[s]' (p.292) that can be compared to living ecologies.Hence, it seeks to investigate the interactions among their elements, both semantic (or strictly linguistic)-definition, examples, usage labels/notes, spelling, pronunciation, etc.-and semiotic (especially in e-dictionaries)-size, medium, interface, voice recordings, pictures, videos, hyperlinks, etc.Moreover, lexicographical resources participate in the "ecology" of distributed cognition, according to which learning results from the interplay among different materials and actors (including lexicographers).The prime objective of micro-ecolexicography is to account for all such variables in the design of effective dictionaries.
Conversely, macro-ecolexicography focusses on the relationship between dictionary compilation and ideology, from an ecological perspective.It considers the role of lexicographical resources in promoting environmental awareness, based on the users' response to available definitions.As such, it draws from the tradition of critical lexicography (Kachru & Kahane, 1995) and Critical Lexicographical Discourse Studies (Chen, 2019).These understand dictionaries as discourse, and accordingly recognise that they 'are not value-free representations' (Liu et al., 2021, p. 298) of the world, but rather 'constitute a repository of the common values and interests of the society whose language is described' (Béjoint, 2010, p. 202).Indeed, cultural data are reflected in dictionaries, whose entries disclose information as to the forms and meanings that the group of speakers represented there 'consider to be (in)appropriate, (un)desirable, (dis)agreeable, or (non)standard' (Łozowski, 2017, p. 166).Macro-ecolexicography is concerned with the ideological elements in dictionaries that reveal dominant stances about how to engage with the non-human world, including natural materials, non-human animals, plants, ecosystems, and so on.Therefore, it involves two types of complementary practices.One is metalexicographical, and consists in the investigation and critique of current entries to evaluate their (mis)alignment with worldviews based on ecological consciousness.The other is genuinely lexicographical, since it comprises the production or rectification of 'dictionary discourse to avoid biased representations of reality' (Liu et al., 2021, p. 305) (Note 2).
Within the history of ecolinguistic research, perhaps one of the first studies to account for the ecological role of words is Trampe's (1991Trampe's ( /2001)).In his work, Trampe reviewed a collection of lexical items retrieved from publications on agribusiness, and identified four main discursive tendencies in the domain: 1) reification, or the representation of living beings as objects or commodities, through nouns and verbs associated with manufactured goods, such as produce, manage, or recipient material (to denote cows or sows to be implanted with an embryo); 2) 'euphemisation', or the use of euphemisms to conceal unpleasant facts about the industry, such as plant protection tool in place of pesticide; 3) denigration of traditional agriculture, described through such items as rural, peasant-like, or retrograde; 4) recurrence of slogans or phraseological elements that advertise intensive farming as inevitable and beneficial, e.g.grow or drop out, ecology or economy (Note 3).Several other studies have followed Trampe's (1991) example, exploring recurrent lexical uses, either in environmental discourse or other domains, and critiquing the ideologies they imply or their effectiveness in raising environmental awareness.These include, for instance, the naming of the climate crisis (Liu & Huang, 2022;Penz, 2018), or the proliferation of compounds to talk about carbon production (Dury, 2008;Koteyko et al., 2010;Nerlich & Koteyko, 2009).However, because they do not consider lexicographical resources, they might be seen to contribute to what Sarmento (2001) called 'ecolexicology'.Heuberger's (2003) is perhaps the study which best exemplifies the aims of ecolexicography.He investigated anthropocentrism in both native speakers' and learners' English dictionaries, searching for entries about non-human animals.He found that, in many cases, definitions presented them with exclusive reference to their usefulness, edibility, beauty, and harmfulness from a human perspective.On the contrary, more objective features such as appearance or typical behaviour were overlooked.For instance, an entry for chicken would read: 'a domestic fowl bred for its flesh or eggs, esp.young' (p.92); or vulture: 'a large ugly bird with an almost featherless head and neck, which feeds on dead animals' (p.97).Therefore, Heuberger (2003) concluded that dictionaries reflected the dominant view shared in most capitalistic societies, which sees non-human animals almost exclusively as resources to be exploited.
A similar trend was observed by Ezzati and Gholinejad Pirbazari (2022) with regard to plants.They reviewed all entries related to plants contained in the Persian dictionary Farahng e Bozorg e Sokhan.Their findings showed that, first, definitions often neglect careful taxonomy and categorise most plants vaguely as either "tree" or "bush"; second, they largely focus on characteristics that are appealing to humans in terms of edibility and/or therapeutic properties, resulting in a distinction between "useful" and "ornamental" plants.Moreover, entries consistently evaluate fruits and flowers by personifying them through adjectives such as "pretty", "ugly", "good-natured", etc.
Taking earlier examples into consideration, the present study tries to expand on the former type of research and directly contribute to Liu et al.'s (2021) macro-ecoloxicography.Although it suggests some possible lines of intervention (see Section 5), it is primarily concerned with the critical stage of the ecolexicographical enterprise.

Methods
This research investigates the noun eco(-)lodge within ecotourism discourse.It aims to 1) explore its meaning according to available definitions and common uses; 2) evaluate its role in contributing to the promotion of responsible travel, and thus discuss the potential of lexical creativity in shaping ecological consciousness.
The methodology draws from a longstanding tradition of cooperation between lexicography and corpus linguistics (Faaß, 2018;Kilgariff, 2022;Rees, 2022).It combines quantitative and qualitative approaches in analysing and comparing data gathered from primary lexicographical resources, i.e., dictionaries, and English corpora.
To begin with, both native speakers' and learners' dictionaries were consulted.The former included the online editions of the Oxford English Dictionary (OED) and the Merriam-Webster Dictionary (MWD), in order to account for both the British and North-American contexts.As for the latter, all the online versions of the "Big Five" were searched-the Oxford Advanced Learner's Dictionary (OALD), the Longman Dictionary of Contemporary English (LDOCE), the Collins Dictionary (COBUILD), the Cambridge Advanced Learner's English Dictionary (CALD), and the Macmillan English Dictionary for Advanced Learners (MEDAL) (Note 4).
Following the retrieval of definitions provided by the abovementioned dictionaries, attention was turned to corpora to investigate examples of usage as found in actual texts.In light of the type of discourse evoked by the noun under scrutiny, the EcoLexicon English Corpus (EEC) (Faber et al., 2014) was initially targeted.The corpus was compiled by the LexiCon research group at the University of Granada.It contains contemporary environmental texts ranging between 1973 and 2016, and covering different domains such as biology, ecology, meteorology, environmental engineering, environmental law, etc.It totals 23.1 million words.After that, and with the aim of exploring the distribution and meaning of eco(-)lodge among non-specialised users, attention was turned to general corpora of English.Three main datasets were accordingly chosen: the British National Corpus 2014 (BNC2014); the Corpus of Contemporary American English (COCA); and, finally, the English Web Corpus 2021 (enTenTen21).In all corpora, all three forms of the lexeme-ecolodge, eco-lodge, and eco lodge-were searched.
As a further methodological step developed to ascertain the frequency with which buildings labelled as eco(-)lodge are qualified with reference to sustainability, the concordances produced by the corpora that showed matches for the search word (see Section 4) were accordingly and thoroughly analysed.In the case of the COCA corpus, it was possible to search through all of them manually.Instead, due to the larger quantity of occurrences found, and in order not to compromise the quality of the examination, the analysis of the enTenTen21 was restricted to a randomly obtained sample of 100 concordances for each form.This allowed to count how many concordances addressed ecological characteristics of the ecolodge, either with regard to the design and construction of the amenity itself or to its management.The following are instances of segments which were marked as befitting the former category:

Example (1)
The extraordinary Costa Rica eco-lodge was designed in harmony with the environment constructed from naturally fallen trees and recycled materials.[enTenTen21] Meanwhile, the ecolodge is a linchpin in the preservation of a 25,000acre conservation area.[COCA] Finally, a collocation analysis was performed to gain insight into the type of items which most often co-occur with eco(-)lodge and therefore participate in its definition.Due to the small number of occurrences in the COCA corpus (see Table 2), this was performed on the enTenTen21 only, using the WordSketch function featured in SketchEngine.Furthermore, as opposed to concordancing, the whole corpus was targeted in this case, and attention focussed on modifiers.

Dictionaries
Some interesting findings emerged from the initial observation that most dictionaries do not include any entry for eco(-)lodge (see Table 1), starting from native speakers' sources.Nevertheless, whereas a search in the MWD produced no result at all, the OED records the hyphenated form eco-lodge.It does not register it as a headword, nor does it provide a definition, but simply lists it among the examples of sense 2.d for the voice -eco-as a combining form, which reads: "Denoting people and things relating to travel to and tourism in areas of ecological interest".It figures at the bottom of an inventory comprising also eco-traveller, eco-adventure, ecotravel, and eco-resort.

Dictionary
Has an entry for eco(- Learners' dictionaries, on the contrary, allowed to obtain formal definitions of the word, albeit only two out of the Big Five include it as an entry.Both the CALD and the COBUILD prefer the form ecolodge, defined as follows: A place for people to stay on holiday that is designed not to harm the environment where it has been built. [CALD] A tourist accommodation facility designed to have minimal impact on the environment, often constructed as part of an environmental project. [COBUILD] As shown, the two definitions indicate minimal environmental impact as the primary, if not only, characteristic denoting this type of amenity.The COBUILD, in particular, suggests involvement with other, non-touristic activities, but retains the focus on ecological protection.The same may be inferred with regard to the OED.
While the latter gives no details as to the kind of elements which qualify the structure as such, it does so indirectly.In fact, with reference to sense 2.d quoted previously, it states that "this activity [tourism to areas of ecological interest] is designed to be favourable to the environment".Therefore, there is a clear indication that the meaning overlaps with sense 2.e of -eco-: "Denoting products, programmes, etc., which promote environmental conservation".Ultimately, according to strictly lexicographical resources, sustainability emerges as the most important aspect of an ecolodge.

Corpora
Figures resulting from the corpora align with the picture obtained from dictionaries, which frames eco(-)lodge as still in the process of being lexicalised (see Table 2).Based on relative frequency, little or no difference is detectable among the four corpora.In fact, in the present case, raw frequencies represent a more relevant value to take into consideration.Provided that the great imbalance observed is mainly due to the very different sizes of the corpora searched (EcoLexicon = 23.1 million words; BNC2014 = 112.1 million words; COCA = 1 billion words; enTenTen21 = 52 billion words), raw numbers still afford the opportunity to make some interesting considerations.The complete absence of eco(-)lodge in the EcoLexicon English corpus is particularly striking.Similarly, no occurrences of either form are found in the BNC2014, as the oldest and least recently updated of the four.The COCA, which collects texts up until 2019, does register some instances-albeit modest.On the contrary, the most conspicuous figures belong to the enTenTen21, comprising texts available online as of January 2022.Based on such a preliminary sketch, it may be speculated that eco(-)lodge is indeed a relatively recent creation, and that it mainly informs online communication as opposed to other media.
The second part of the analysis moved to the examination of concordances in order to assess how frequently discussions of ecolodges co-occur with mentions of their ecological impact.At a first glance, the comparison reveals that while most of the occurrences in the COCA contain references to sustainability, they do not in the enTenTen21.In spite of this, results gained from the former must be interpreted in light of the representativeness afforded by the sample under consideration.The 28 hits obtained for ecolodge*, for example, appear in 7 texts only.Out of the 25 instances in which ecological matters are addressed, as many as 18 are featured within the same article, titled "The eco home"; 5 belong to an environmental magazine; and 1 appears in a contribution to an academic journal discussing forest preservation.Thus, results are influenced by the fact that most of the COCA instances are documented in texts with an explicit environmental focus.
Instead, the enTenTen21 shows a considerable low frequency of mentions of ecological features in concordances for eco(-)lodge.Based on such preliminary findings, it may be argued that while sustainability is certainly accounted for in descriptions of such amenities, it is not as significant a factor as expected.Indeed, the substantial majority of the concordances abstains from referring to it.In addition, even though the enTenTen21 does not provide a topic tag for all the texts it includes, distribution still highlights that, as opposed to the COCA, here most of the texts recording the word belong to the domain of travel and tourism (see Table 4).With the aim of gaining better understanding of the contextual items that typically co-occur with eco(-)lodge, a collocation analysis was carried out at this point.The list of modifiers obtained for each form is particularly revealing (see Table 5).Some "noise" or interference is produced here by elements that are used in the proper names of facilities mentioned multiple times within the corpus.For instance, Sina and Selva are part of "Mt.Sina Ecolodge" and "Selva Verde eco lodge".Also, Rica features in "Costa Rica ecolodge/eco lodge", since the corpus contains a few texts that specifically review and advertise different accommodations in this destination.These items were disregarded in the analysis, as it was concerned with modifiers containing information about the characteristics of ecolodges.
The remaining items are indeed more salient.Three main semantic classes of words may be identified, qualifying the ecolodge according to its location, environmental impact, and level of comfort.The first group is composed of topographical modifiers, mostly nouns, including rainforest-which is the most consistent and characterises all three forms-riverside, safari, beachfront, and jungle.

Example (2)
In Panama City, glass towers and boutique hotels dominate, though a drive just outside the city reveals a world of rainforest ecolodges, thatched beach huts, bird-watching resorts, and coffee haciendas.
Your second base is a remote, riverside eco-lodge where you can bird-watch, hike, horseback ride, or explore the rainforest canopy with your local guide.
The second group comprises collocates that pertain to the domain of sustainability.It registers only two instances: off-grid, which denotes the absence of connections to public utilities or power services like electricity or gas; and solar-powered.

Example (3)
The off-grid eco lodge has been designed to maximise the panoramic views and harness the power of the sun.
A stay in this true solar-powered eco-lodge in the wilderness, about an hour from Brus Laguna, costs just $10 a night.
Finally, the remaining group of modifiers refers to the ranking of the amenity, or rather its degree of opulence.
Apart from rustic, they all suggest high-end status.Both the nominal and adjectival forms luxury and luxurious are included (together, they collocate 124 times with ecolodge), followed by boutique, upscale, and five-star.

Example (5)
Starting and ending in San José, the two of you will ford the scenic Río Pacuare (one of the best white-water rafting rivers in the world), explore lush jungle, and enjoy downtime together in a luxury eco-lodge.
Your stay at this luxurious eco-lodge includes gourmet Ecuadorian meals, non-alcoholic drinks, and all activities and guided tours.
House in the Wild is a private boutique eco lodge, tucked away on a 1000 acre private estate, Naretoi, on the edge of the Maasai Mara.
Frégate features a luxurious five-star eco-lodge offering the optimum in comfort and amenities that has become a favourite hideaway for Hollywood stars, with deluxe villas right on the foreshore to ensure each has a million-dollar sea view.
Hence, the collocation analysis seems to indicate that, for the most part, ecolodges are described as being built in non-urban contexts, especially in natural areas conceived of as wild or untamed, such as forests or jungles.As a matter of fact, no occurrences of the noun eco(-)lodge denoting amenities built in urban contexts were found.Their ecological impact is rarely emphasised.In contrast, they are often presented as exclusive and expensive types of accommodation, catering to a restricted target of well-off tourists.
Furthermore, a closer look at concordances for these collocates, reveals that comfort and ecology are mostly kept separate, so that when the luxury aspect of ecolodges is addressed, references to sustainability are less frequent.
When both features do appear, the latter are actually associated with other amenities located in the same area as the ecolodge, like in the following examples:

Example (6)
We stayed in luxury, [sic] eco-lodges and a house made entirely of mud with no electricity and walls open to the elements.
The striking Malewa Wildlife Lodge […] has been transformed into a stunning luxury eco-lodge […] Only 1km from Malewa Wildlife Lodge, Kigio Wildlife Camp is a superb all-suite camp […] constructed from sustainable pine-wood, local earth from the conservancy, traditional thatch provided by the surrounding community and canvas panels-there has been minimal usage of cement and steel and the environment has been carefully preserved during its construction.
Also, seen against the productivity of the category of opulence, the adjective rustic stands out in the collocates list, and opposes the main idea of retreat to luxurious ecolodges.The presence of both rustic and luxurious would suggest that the abodes have no standards in terms of ranking, and may offer either sumptuous and costly stays or humble and economical ones.Again, the qualitative examination of concordances for the former item as a modifier of ecolodge and eco-lodge allowed to clarify such apparent mismatch.The following example shows that rustic is not interpreted as "modest", "rough", "plain", or "simple":

Example (7)
Perfect for families, you'll also find on the property a giant tree to climb, gym equipment, a tennis and basketball court, plus an advertised heated pool that is often too cold to use, possibly due to the cooler climate.Onsite spa services are available.It may be noted that this rustic eco-lodge offers more in the way of amenities than most other available hotels in the area.
In fact, "rustic ecolodges" offer as many first-class services as luxury ones, and are equipped with the same kinds of amenities, such as gyms, basketball courts, heated pools, etc. Rather, the semantic value of rustic seems to concern the location of the ecolodge, and is constructed negatively in opposition with "urban".That is, the main meaning of rustic in the examples is actually "far from town", and therefore the word may be included in the group of modifiers referring to the positioning of such accommodations in natural areas.
This reinforces the idea according to which "true" nature can be experienced solely by getting away from urban settlements and into the wilderness surrounding them, i.e., their literal "environment".Such a tendency is observable in yet another modifier collocating with eco(-)lodge, namely, secluded.In point of fact, whereas ecolodges are built with the aim of granting closeness and direct access to nature, the former adjective qualifies the structures as separate-perhaps even shielded-from the places they are found in.Therefore, proximity to nature is mediated by means of deliberate confinement, precisely because the former is perceived as untamed, and hence potentially dangerous.A possible explanation for the insistence on the great comfort of ecolodges in these texts is that luxury is offered as an adequate and necessary compensation for the uncomfortableness of travelling to such distant, wild, and perilous destinations.

Example (8)
Enjoy the clean air, peace and tranquility at this secluded eco-lodge.You will lay in a hammock and listen to birds sing while being served refreshments.
Luna Lodge is a secluded eco-lodge set in the pristine and captivating Osa Peninsula, near Corcovado National Park in southwestern Costa Rica.

The 'Eco-' in 'Eco(-)lodge'
The analysis of dictionaries and corpora highlights a discrepancy between definitions and uses of eco(-)lodge.
Both sources identify the structures as tourist facilities.However, on the one hand, lexicographical entries present them as having low environmental impact, and qualify ecolodges with exclusive reference to ecological consideration.On the other hand, instances retrieved from texts tend to overlook issues of sustainability, and mainly characterise the buildings as accommodations for five-star sojourns in natural areas.
One possible reason for such a shift may be ascribed to the polysemy of the adjective ecological itself, which has currently two main meanings.The first-and oldest, dating back to Ernst Haeckel's coinage of the term in German at the end of the nineteenth century-relates to biology, or rather ecology as a science.In this case, it indicates involvement with the interrelationships among organisms, and between organisms and their habitat.
The second meaning is more recent-the OED's first entry dates back to 1969-and it stands as a synonym of environmental, that is, concerned with the impact (and damage) of human activity on the natural environment.
Both senses are potentially carried by the combining form eco-, which has been particularly productive as a prefix in the creation of new words since the beginning of the present millennium (Benz, 2001).However, it would seem that the proliferation of "eco-words" corresponds to a gradual semantic opacification of the pre-modifier.Dury (2008) registered such phenomenon as an instance of Meyer and Mackintosh's (2001) 'de-terminologization', the process by which grammatical, semantic, and pragmatic shifts may occur when non-specialists borrow terms from a specialised domain that has become of interest to the general public.In the case of ecology, the recognition of the importance of ecosystemic relationships in the well-being of the natural environment has led non-scientists and laypeople in general to adopt eco-to denote products and attitudes mindful of ecological principles, regardless of their actual link with the discipline.According to Dury (2008), this is observable in the fact that "eco-terms" that have spread in general language are not equally used in specialised language.
The flexibility of eco-has made the former highly creative in ecotourism discourse (Raţă et al., 2012).Not only has it provided the activity with its name, but it has also inspired a rich set of lexical innovations to describe its actors (e.g., eco traveller), activities (e.g., eco trekking), and destinations (e.g., eco camp).These uses do not pertain to the science of ecology.However, they do not strictly convey the sense of "environmentally-friendly" observed in the combining form's de-terminologization. Indeed, the case of eco(-)lodge seems to testify to a further semantic shift in the meaning of eco-.As observed by the investigation of the corpora, and especially in light of the collocation analysis performed, the prefix does not necessarily suggest ecological concerns in the construction and maintenance of the facility, but rather refers to its location in non-urban environments (e.g., "rainforest ecolodge").These findings confirm Mühlhäusler and Peace's (2001) early assumption about the 'severe semantic bleaching' of eco-, which '[r]ather than referring to functional interrelationships between the inhabitants of an ecology, […] has come to mean something like "having to do with nature''' (p.378).Such "general" connection with nature is ultimately only topographical.Provided that is the case, pre-modification through eco-would be tautological, since ecotourism, by definition, already implies (sustainable) activities that specifically take place in non-urban contexts.
Moreover, this movement from "environmental" to "in natural environments" is documented by the textual distribution of eco(-)lodge noted in the present study.Occurrences of the lemma in the COCA mostly contain explicit reference to sustainability, and these are found in environmental texts which seem to describe in great detail the ecological-i.e., environmental-features of the buildings under consideration.In contrast, occurrences listed in the enTenTen21, which rarely address ecological matters explicitly, belong to (eco)tourism discourse.
It is possible that ecotourism texts exploit the semantic ambiguity of eco-to advertise ecolodges and cater to conscious travellers.Because the combining form has by now taken the meaning of "opposing damage to the environment", and it is this latter sense that has prompted most lexical creations in recent years, readers may assume that the sustainability of the accommodation is inherent in the very use of the noun eco(-)lodge.In part, this could justify the low frequency of "green concordances" and collocates pertaining to the domain of environmentalism that was observed in the enTenTen21.However, previous studies about sustainable tourism found that sustainability tends to be consistently promoted in discourse, and related lexical items often register as keywords (Malavasi, 2017).Instead, the present analysis has revealed that, at least as far as ecolodges are concerned, comfort, luxury, and (secluded) proximity to natural areas are prioritised over environmental impact.Benz (2001) discussed the semantic transformation of eco-from scientific to general use in terms of politicisation, considering how vocabulary was being re-shaped so as to meet the growing awareness about and campaigning for the state of the natural environment.Accordingly, he concluded his review of lexicographical sources by stating that 'eco(-) plays a significant role in the lexicon of current English, proof that green discourse is an important register of current English, proof that there are increasingly diverse ways to speak and write of the environment and care for the earth' (Benz, 2001, p. 169) (Note 5).As opposed to this, present trends observed in the discursive construction of eco(-)lodge point to a reverse process, that is, a de-politicisation of eco-, whose "environmental" sense is being tempered to vaguely mean "found in natural, non-urban areas".

Conclusions
The analysis of eco(-)lodge has revealed that the newly-formed combination has not reached complete lexicalisation yet.To begin with, most of dictionaries do not include it as headword or, as is the case with the OED, only register it as an example.Second, no agreement is found as to the main orthographical form, as eco lodge, eco-lodge, and ecolodge are all documented (with a slight preference for the former two variants), and hint at different degrees of lexicalisation.Finally, a partial semantic discrepancy between definitions and uses is observed.Dictionaries qualify ecolodges with exclusive reference to their environmental impact, whereas discursive instances often overlook the former and instead emphasise location and level of comfort.
Possibly, the gap between dictionary entries and corpora may be explained chronologically.Considering the differences that emerge in terms of textual distribution between the COCA and the enTenTen21, with the latter being more up-to-date, it is possible that dictionaries might have drawn their data from earlier occurrences of the noun.In other words, prior to becoming predominant in ecotourism advertising, eco(-)lodge might have been used chiefly in environmental discourse.A sign of this may be found in the OED, which includes the following example of usage for eco-lodge, dated 1991: 'By next year they plan to have an ecolodge and research station that will include walkways suspended in the canopy of the rainforest to be used by tourists and researchers'.This instance seems to indicate a double function for the ecolodge: it hosts both tourists and scientists engaged in doing research in the same area where the structure is built (similar to what the CALD entry implies, see Section 4).For this reason, it is likely that the term came to be restricted to the field of tourism somewhat later.However, a further and systematic diachronic analysis of eco(-)lodge is needed in order to test this hypothesis.
Still, all the dictionaries examined register a specialised use of the noun, denoting a type of tourist accommodation.As available entries bind the definition to the facility's environmental impact only, they do not acknowledge the meanings retrieved from the investigation of the corpora.On the one hand, they might be seen as misleading, since they induce users-learners, in particular-to assume that all ecolodges are sustainable "by definition".As a result, they might fail in uncovering promotional tourism texts (and their producers), which greenwash their activities by exploiting eco(-)lodge as a catchword.Within ecolexicography, in case of controversial entries, or entries which support an unfavourable representation of reality, especially in terms of ecological consciousness, Liu et al. (2021) recommend direct intervention through the modification or update of the former.They suggest five main ways of 'reframing the dictionary discourse: warning, commenting, refining, questioning and neutralizing' (p.305).In this case, lexicographers could intervene by adding modal items-e.g."generally", or "in most cases", placed at the beginning of the definition-to allow for exceptions.They could also add further illustrative examples, given their 'potential [as] lexicographic resources of cultural data' (Łozowski, 2017, p. 166).Examples would avert such "neutral" sentences as, for instance, 'I'm currently building an ecolodge' [COBUILD], allowing for others of a more evaluative character-e.g., 'You can pay £3,000 to stay in a luxury ecolodge' [CALD].
On the other hand, definitions of eco(-)lodge might be considered as being acceptable, regardless of the factual sustainability of the structures addressed as such.Accordingly, the distance between lexical entry and use may be preserved so that the former represents a "standard", enlisting the properties a building must possess in order to be considered an ecolodge.This aligns with Rees's (2022) comment on the interactions between dictionary and corpus, as he observed that corpus-based and corpus-driven approaches to lexicography should face the fact that a dictionary is 'often regarded as a repository of "correct" language use' (p.391).
In conclusion, what is more revealing about the present analysis, is that it addresses some (linguistic) shortcomings of the ecotourism industry.Admittedly, comfort and sustainability are not mutually exclusive, and an ecolodge could be built to be both comfortable and sustainable.As a consequence, promotional materials will advertise environmental impact and other factors, such as ranking of the structure, services provided, and so on.Stamou andParaskevopoulos (2003, 2004) understand ecotourism discourse as instantiating itself along a continuum which has tourism (i.e., economy) and environmentalism at its ends, so that elements of both discourses will always inform ecotourism texts.Results from the current research, however, register an imbalance between the two.In fact, discursive constructions emphasise luxury over sustainability, to the extent that the eco-in eco(-)lodge loses its environmental connotation and refers only to the location of the amenity.
Ultimately, the present study demonstrates how ecolexicography-intended not only as the production of lexicographical materials about ecological vocabulary, but rather as the critical-discursive examination of texts which include lexicographical resources-can be used to investigate the interrelationship between language and ecology, and assess the viability of its outcomes, particularly in terms of lexical innovation.

Table 2 .
Occurrences of eco(-)lodge in the selected corpora

Table 4 .
Distribution of eco(-)lodge in the enTenTen21 according to topic