Revisiting the Description of Tense in English

The controversy surrounding the description of tense in English has remained because scholars have concentrated on carving descriptive niches for themselves rather than paying appropriate attention to its causative factor(s), resulting in three different descriptions: traditional, structural, and systemic. This paper identifies the genesis of the problems, points out how this hinders the attainment of descriptive accuracy, and proffers some solutions. It contends that arguments, such as whether or not there is a future tense for English, stem from the way tense is generally conceptualised. It examined ten standard definitions of tense and found that the keyword grammaticalisation is narrowly interpreted to mean the morphological only, whereas a language’s grammatical system consists of both syntactical and morphological aspects. The non-recognition of the syntactical component—even by grammarians that acknowledge future tense—is the root of the descriptive issues with tense. The paper proposes syntactical marking, achieved by placing the auxiliary WILL/SHALL or BE GOING TO before the base form verb, as the mechanism for future tense marking in English. In effect, English has a three-tense system, and its modes of marking are morphological (for present and past tenses) and syntactical (for future tense).


Background
The title of this paper suggests at least two things about tense in English, namely that it has been thoroughly described in the grammar of English and that there are still outstanding issues despite the enormous literature on the subject (These are long-acknowledged facts). Tense is controversial and problematic to describe, and it is no less so for English (Declerck, 1991;Depraetere & Salkie, 2015). For instance, Lyons (1968) notes that "the analysis of tense even in English is a matter of considerable controversy" (p. 306), while Comrie (1985) similarly observes that "even for a language as thoroughly researched as English there remains controversy concerning the definition of the various tenses" (p. 9). The major areas of disagreement seem to be on (1) whether there are two or three tenses in English, (2) whether or not there is a future tense marked by WILL/SHALL, and (3) whether or not tense selection is made more than once in a given verbal group (Palmer, 1987, pp. 37−38;Palmer, 1990, pp. 12−13). It is this controversy surrounding the description of tense in English that resulted in three different descriptions of a single linguistic category, with each one claiming greater descriptive accuracy over the others. These are the traditional, structural, and systemic descriptions (Aremo, 1984). For a grammatical system as important as tense in English controversies are not unexpected, but I do not intend to delve much into the arguments beyond what I have just hinted at here and outside occasional remarks (which is, merely acknowledging that they exist) for two reasons. The arguments have been so exhaustively thrashed out that discussing them further adds little or no value. Besides, doing so will take me off-course, which is to identify and account for the root of the controversy associated with the description of tense in English. This paper does not therefore seek to rehash existing descriptions of tense in English; neither does it seek to simply catalogue what are traditionally regarded as the problem areas associated with them. These are well documented elsewhere (Adejare, 2014). What it does seek to achieve is to identify the cause of the issues that there are, and relate this to specific aspects of the theory and description of tense as a linguistic category.

The Problem
have concentrated on carving descriptive niches for themselves rather than paying appropriate attention to its causative factor(s). It is my contention that the controversy stems from the way tense itself is generally conceptualised, particularly the narrow interpretation of the keyword in some of its notable definitions, namely grammaticalisation. No study is known to have examined the problems of tense in English from this perspective.

The Purpose
This revisit of the description of tense in English is to serve the following purpose: (1) To identify the root cause of the problems associated with the description of tense in English (2) To demonstrate that the term grammaticalisation is restrictively applied to the morphological aspect only, causing the unwarranted argument against future tense and its non-morphological markers WILL/SHALL and BE GOING TO (3) To prove that the mechanism for future tense marking in English is syntactical, and subsequently make a case for its recognition in addition to the already established morphological marking for past tense and present tense (4) To highlight some implications of recognising syntactical and morphological modes of marking tense in English for both theory and description.
At this point, it is necessary to provide an insight into the structure of the paper and how the rest of it will proceed. Immediately following this introduction is a brief discussion of the method adopted for the study, after which is the review of literature. Section 4 articulates the main argument while the fifth and final section concludes the paper.

Method
The nature of the topic, coupled with the purpose for exploring it, does not require the collection of field data. In this regard, materials are critically reviewed in order to properly situate the main argument against existing descriptions in relation to the research aims. The data consists mainly of ten standard definitions of tense extracted from the literature. These are evaluated in support of the arguments advanced. The rest comprises examples already cited in the literature, used to affirm, reinforce, or counter relevant descriptive statements already in existence. The reuse of data from existing works is not new in linguistics (cf. Sarkar, 1998).

Literature Review
This review of literature centres on the conceptualisation of tense in the grammar, and it naturally includes its relationship with time and its definitions. Because of the centrality of the concept grammaticalisation to the main argument of this paper, the review begins with its explication. The theoretical framework adopted for the study ends the review.

Grammaticalisation
There are two senses in which the term grammaticalisation is used in linguistics, according to Hopper and Traugott (2003). The first is with reference to a research framework within which phenomena of language are accounted for. Here, it "refers to that part of the study of language change that is concerned with such questions as how lexical items and constructions come in certain linguistic contexts to serve grammatical functions or how grammatical items develop new functions" (p. 1). The second meaning refers to, not the study of the phenomena, but the phenomena themselves; that is, it refers to the change. Thus, grammaticalisation is defined as "the change whereby lexical forms and constructions serve grammatical functions in certain linguistic contexts, and once grammaticalized, continue to develop new grammatical functions" (p. 1). Kiparsky (2012) points out two competing definitions of this second sense of the term in new functionalist approaches, where the divergence is seen in the nature of the change itself: Is it "new" or "more" grammatical functions? On this issue, Kisparsky affirms Hodder and Traugott's definition by stating that grammaticalisation is "a change by which an element acquires 'new grammatical functions' (rather than 'more grammatical functions')". For him, it is a concept "effectively equated with grammatical (morphosyntactic) change" (pp. 4−7). It is this second meaning of the term that is pertinent to the current study. Grammaticalisation thus means "integration into the grammatical system of a language", and this is opposed to lexicalisation, which means "integration into the lexicon of the language" (Comrie, 1985, p. 10).
The idea of grammaticalisation as integration into a language's grammatical system needs to be put into its proper perspective because the concept of grammar is crucial to grammaticalisation. It requires linking to grammar itself, digging somewhat deep into grammatical theory, even if only briefly. The linking goes thus. There are three primary levels of linguistic description recognised in Systemic Grammar, which are substance, form and context. The level of form, which organises the recurring internal patterns of a language into (3) "Tense (adapted from the OED) is any one of the forms in the conjugation of a verb which serves to indicate the different times at which the 'action' is viewed as happening or existing." (Strang, 1969, p. 143) (4) "Tense…grammaticalises the relationship which holds between the time of the situation that is being described and the temporal zero-point of the deictic context." (Lyons, 1977, p. 678) (5) "Tense is grammaticalised expression of location in time". (Comrie, 1985, p. 9) (6) "…we prefer to follow those grammarians who have treated tense strictly as a category realised by verb inflections…" (Quirk et al., 1985, p. 176) (7) "…we prefer to limit …the term tense to the morphological opposition between present and past forms of the finite verb." (Quirk et al., 1985, p. 189) (8) "Tense is a grammatical category referring to the location of a situation in time. Strictly speaking, English has two tenses of the verb-present and past-if tense is defined as being shown by a verb inflection." (Greenbaum, 1996, p. 253) (9) "Tense…is a grammaticalization of temporal relations. This means among other things that it is realised through auxiliaries and that these verbs are much more generalised than their lexical homonyms." (Matthiessen, 1996, p. 10) (10) "…the term tense is used to indicate the tense morphology of a language which refers (not exclusively though) to the temporal representation. This notion is used to denote the grammatical category realised by the inflectional elements. This is tense or tense morphology." (Sarkar, 1998, p. 92) The definitions reveal two facts about the way in which tense is fundamentally conceptualised. Semantically tense is intrinsically associated with the notion of time, so much so that the word time occurs in eight out of the ten (eighty percent) definitions of tense cited (The only exceptions are 6 and 7). This confirms the strongly held view that there can be no consideration of tense without reference to time, which makes tense truly the linguistic category for handling "time relations". Descriptively tense belongs essentially to the level of linguistic form called grammar. All ten definitions (without exception) point to the fact that tense is a grammatical category, and this is instructive. So, tense is a category of grammar that serves to make reference to time. It is thus a grammatical category as well as a category of meaning.
Placed within the context of the concept grammaticalisation examined above, it would be seen that the definitions (and the analyses they lead up to in the source texts) reveal narrowness in the interpretation of the term grammaticalisation and its "synonyms". First, thirty percent of the definitions (4, 5 and 9) expressly contain the term grammaticalisation, while seventy percent (1, 2, 3, 6, 7, 8, and 10) contain the "synonyms" morphology, verb form, inflection, tense morphology, and systematic grammatical contrasts. Second, only one definition (9) makes reference to auxiliaries, which impliedly admits that grammaticalisation is not limited to verb morphology and that it can also involve auxiliary verbs. Definition (9) is however silent on how the auxiliary operates and on the other means of grammaticalising temporal locations. Ninety percent of the definitions overtly and unambiguously equate grammaticalisation with verb inflection or verb morphology. This means that only the morphological component of the grammatical system is acknowledged as responsible for the linguistic expression of location in time, to the total exclusion of the syntactical component. The phrases "verb forms", "systematic grammatical contrasts", "conjugation of a verb", "tense morphology", and "verb inflections" expressly attest to this conclusion. Here lies the genesis of the problems surrounding the accurate description of tense in English.

Tense as the Grammaticalised Expression of Location in Time?
In the light of the foregoing revelations, I submit that grammaticalisation has been unduly, narrowly interpreted to mean the morphological, rather than cover both the morphological and the syntactical, and that this restricted interpretation is the root of the issues associated with the description of tense in English. The limited application of the concept grammaticalisation particularly blurs the proponents of the two-term system of tense in English from recognising future tense, its markers, and mode of marking. There is even an indication that some proponents of a two-term system for English admit the possibility of tense marking in ways other than the morphological. Definitions (6)−(8) are proofs. But why would grammarians working on tense deliberately choose to uphold only one component of a concept and present it as the whole? The reason is not far-fetched with respect to the current subject matter. The syntactical component of the grammatical system of a language lacks explicit indicators, and is therefore not easily recognisable or identifiable particularly because of its chain-like nature. In contrast, its morphological counterpart is overtly associated with forms of the verb lexeme, which makes it catchy. Descriptive atomism is not new in, or strange to, linguistics after all. It has been ijel.ccsenet.org International Journal of English Linguistics Vol. 12, No. 4;2022 demonstrated that, right from the time of Ferdinand de Saussure, linguists have mainly concerned themselves with FORM to the total exclusion of SITUATION, whereas FORM and SITUATION are the twin obligatory elements of structure of language, and that that explains why an accurate account of natural language is lacking to date (O. Adejare, forthcoming). In summary, there are two sides to grammaticalisation (the morphological and the syntactical aspects), and both must be recognised in order for an accurate and full account of tense as it exists in actual language use situations to be made.

Theoretical Foundation
The theory of tense upon which this study is anchored is the systemic grammatical theory expounded by Michael A. K. Halliday. Systemic Grammar recognises three tenses for English (present, past and future) in accordance with the three-time phases (present, past and future) identifiable in all cultures and in tandem with the traditional description. Future tense is marked by WILL and SHALL in addition to GOING TO (e.g., will dance), as opposed to the mechanism of verb inflection applicable to present (e.g., dance) and past (danced), as espoused by structuralists. The argument advanced is that "all languages show such morphological irregularities". Systemic Grammar uniquely argues that tense is recursive (That is, tense selection is made more than once in a structurally compound or complex verbal group). It does not however distinguish between tense and aspect (Halliday, 1961(Halliday, , 1976(Halliday, , p. 149, 1985Halliday & Matthiessen, 2004). In a paper aimed at locating Halliday's interpretation of tense on the overall interpretation of English, Matthiessen (1996) reviews arguments against future tense and identifies two interpretations of tense termed the reductionist and expansionist interpretations. The reductionist interpretation of tense is based on a morphological interpretation of tense as a word rank, which reduces the three-term system of past/present/future to a two-term system of past/present on the argument that, while there is a past suffix, there is no future suffix. Scholars cited as espousing this view are Priestley (1768), Joos (1964) and Palmer (1974). The expanded model of tense, in addition to the three tenses of past/present/future, further accounts for compound tenses in terms of primary versus secondary. It brings into focus, not only forms of single verbs but chains of verbs to build a serial model, clearly breaking from the traditional linear model of tense. Thus, Halliday's interpretation of tense within the systemic-functional approach locates tense at the rank of group rather than at the rank of word.
I espouse the systemic description of tense because of its capacity to account for all naturally occurring English tenses. Tense is a system of the verbal group, which is an exponent of the median grammatical unit group. English has five hierarchically ordered units of grammatical description named the sentence, clause, group, word, and morpheme. The group is made up of one or more words. The verbal group may thus comprise one word only (e.g., the finite verb, go, in I go to school every day), two words (e, g., will go in I will go to school tomorrow), or more than two words (e.g., have been going in I have been going to school every day since the new term). The two-or-more word verbal group necessarily requires the placement of one or more auxiliary verbs before the non-finite lexical verb. This explains why tense is said to operate at the rank of group rather than at the rank of word (Halliday, 1976;Matthiessen, 1996;Adejare, 2013).
Though the systemic description recognises a future tense in English marked by WILL and SHALL in addition to GOING TO, its explanation for the auxiliary markers of future tense is unsatisfactory. It is not sufficient to argue dismissively that "the mechanism for producing a future tense marking is different from that used in producing present and past" and that "all languages show such morphological irregularities" (Halliday, 1976, p. 149). Such an argument must be supported with evidence specifying the mechanism involved. This is yet another motivating factor for this revisit of the description of tense in English.

Fallouts of Restricted Conceptualisation of Grammaticalisation
The earlier statement made to the effect that the controversy surrounding the description of tense in English is traceable to how tense is conceptualised led to a critical review of the term grammaticalisation and some notable definitions of tense in the grammar. It was noted that there are two arms of grammaticalisation (morphological and syntactical) and that the term and its close "synonyms" recurred in the definitions of tense sampled (e.g., "Tense is the grammaticalised expression of location in time"). It was also noted that its meaning-integration into the grammatical system of a language-was restrictedly interpreted in the literature to mean morphological integration only and exclude syntactical integration. It was then concluded that this restricted interpretation of tense is the root of the problems associated with its description. This limited interpretation has made it difficult to accurately account for tense in English in all its ramifications, particularly the question of a future tense system, even by scholars who recognise it. In this section, I shall examine what I term the fallouts of the argument that the restricted interpretation of grammaticalisation, which inaccurately excludes the syntactical arm of grammar, leads to an account of tense in English that does not reflect the everyday reality of natural language use. In more specific terms, I shall contend that the restriction of grammaticalisation to the morphological, rather than an embrace of both the morphological and the syntactical (the two component parts of grammar), prevents the recognition of future tense as a term in the system of tense in its own right. As a consequence of this first argument, I shall further argue that syntactical marking is what obtains in respect of future tense in English and therefore make a case for two modes of marking tense, morphological and syntactical. Finally, I shall examine the status of WILL, BE GOING TO and the to-x+o non-finite form as markers of futurity at different levels of delicacy (These also attract controversy in their own unique ways).

Future Tense System in English: A Case for Syntactical Marking
There is not much to be gained by repeating the arguments for and against the existence of a future tense form for English, since these are well documented in the literature and have been hinted at earlier. However, it is useful to restate that there exist three descriptions of tense in English, namely the traditional, structural and systemic (Aremo 1984), and that all three descriptions recognise past tense and present tense marked by verb morphology. However, only the traditional and systemic descriptions recognise future tense marked by the auxiliaries WILL and SHALL, with the latter recognising GOING TO in addition. I subscribe to the systemic view that English has a future tense marked by WILL/ SHALL and BE GOING TO (not GOING TO as will be made clear soon), but I feel strongly against the absence of any clear-cut indication of the mode of marking future tense. In the paragraphs that follow, I shall argue with examples that the narrow interpretation of grammaticalisation, which excludes the syntactical arm of grammar, leads to an account of tense in English that does not accurately reflect the fact of future tense. It is my conviction that, if we recognise three time-phases (including future), there should be a way of grammatically accounting for future time reference particularly in a language that already has tense.
The strongest arguments against future tense (which are that it is not indicated by verb inflection and that the traditional future tense markers are more modal-like than tense-like) stem from the limited interpretation of the concept grammaticalisation pointed out above. The failure to explore the syntactical aspect of grammaticalisation in addition to the morphological resulted in the non-recognition of future tense as a term in the system of tense in its own right. It is therefore necessary to begin this examination of the theoretical and descriptive fallouts of that restriction by identifying the mode of marking future tense: This is syntactical marking. Future tense in English is marked syntactically by placing the auxiliary verb markers before the base form lexical verb to form a chain-like sequence. In other words, this occurs at the syntagmatic axis (axis of chain), as opposed to past tense marking, which takes place at the paradigmatic axis (axis of choice). The following sentences taken from the literature illustrate the syntactical marking of future tense in English. The group boundaries are demarcated and the verbal groups are italicised to further demonstrate the grammatical fact of syntactical marking of future tense.
(1) ||John| will leave |tomorrow||. (Comrie, 1985, p. 46) (2) ||I |'m going to give |a paper| next Wednesday. || (Palmer, 1987, p. 38) (3) ||I| shall give |a paper | next Wednesday||. (Palmer, 1987, p. 38) (4) |Will| he| come? || (Palmer, 1987, p. 38) (5) ||I| will be |perfectively frank | with you||. (Palmer, 1990, p. 160) (6) |||Well| if |I |get bored |with the company ||I | shall come|| and || find |you. ||| (Greenbaum, 1996, p.259) It is important to state at this juncture that syntactical marking of a system, or term thereof, is not new to the grammar of English. However, it would be simplistic to argue that because syntactical marking is the norm elsewhere, it should be valid for future tense. Syntactical marking is what obtains in respect of aspect, the system most semantically related to tense. Progressive aspect and perfective aspect are marked syntactically by placing the appropriate form of BE or HAVE auxiliary before the x+g or x+n form of the non-finite lexical verb as in "he is examining" / "he has examined" (Quirk et al., 1985, p. 189). This grammatical fact is affirmed by Quirk et al. (1985) thus: In fact, aspect is so closely connected in meaning with tense, that the distinction in English grammar between tense and aspect is a little more than terminological convenience which helps us to separate in our minds two different kinds of realization: the morphological realization of tense and the syntactic realization of aspect (p. 189).
Once again, the problem caused by the narrow interpretation of grammaticalisation to cover only the morphological aspect is brought to the fore. The phrases "morphological realization of tense" and "syntactic realization of aspect", presented as the main distinguishing features between tense and aspect, are descriptively ijel.ccsenet.org International Journal of English Linguistics Vol. 12, No. 4;2022 misleading. Morphological marking applies only to present and past tenses whereas syntactical marking applies to aspect and to tense through future tense, as the foregoing discussion shows. In fact, the difference between tense and aspect has shrunken while the affinity between them has widened (if not deepened), with the establishment of syntactical marking for future tense in English. The relationship between tense and aspect, which makes them combine freely in the complex verbal group, now extends beyond the acclaimed semantic ties to include the hitherto unexplored syntactic closeness: the syntactical marking of future tense and the syntactical marking of progressive aspect and perfective aspect.
The question could also be raised as to why the syntactical marking of future tense did not attract attention over the years despite the acknowledgement of a future tense system for English marked by WILL/SHALL Heine's (2017) observation on TG may provide an answer. According to him, syntax was backgrounded in the study of grammaticalisation probably because of the revolt against generative syntax of the 1960s, which effectively equated language study with syntax. This could also explain why grammaticalisation only references morphology, instead of both morphology and syntax.
It is thus clear from the foregoing that future tense in English is marked syntactically by placing the modal auxiliaries WILL/SHALL, or the semi auxiliary verbs with quasi modal functions BE GOING TO, BE ABOUT TO or BE TO, before the x+o base form lexical verb. In effect, there are two grammatically distinct modes of marking tense in English: morphological distinction (for present and past tenses) and syntactical sequencing (for future tense). Having established that future tense in English is marked syntactically by placing WILL or its allies before the x+o base form verb, it becomes necessary to examine the status of WILL as the primary marker of future tense. This is significant in view of the argument that WILL is a modal auxiliary and not a tense auxiliary.

WILL as a Marker of Future Tense
A chief opponent of a future tense system for English marked by WILL is Rodney Huddleston. He is so insistent that WILL is not a marker of future tense but a marker of modality that, even when it is evident that both meanings converge on a given verbal group, he whittles down that of tense by stating that, "where… the modality component is least apparent, there is evidence that it is not entirely lacking". Huddleston vehemently opposes a grammatical split between modal WILL and future tense WILL, but admits that there is difficulty drawing "any reasonable clear boundary between an allegedly purely temporal use and the others" (Huddleston, 1984, p. 174;Huddleston, 1995). Sarkar (1998, p. 113) probes into the question of whether WILL is a future tense marker or a modal auxiliary and concludes, after examining various analyses of WILL and reviewing every argument for and against, that "each instance of will is simultaneously a modal and a tense morpheme". Michaelis (2020) surveys English tenses and tense uses and comes to the conclusion that the tense system is based on the opposition between past and present (which effectively rules out future tense), noting that English has no morphological future tense but a periphrastic construction containing WILL. Lehmann (2002, p. 26) points out that the future may arise through the grammaticalisation of "desiderative modal", and cites English WILL as a known example. Depraetere and Salkie (2015) describe WILL as "a tense marker less grammaticalized than inflection" (p. 358). In their review of the linguistic realisations of future time reference, Jӓggi et al. (2020) classify languages into three according to whether they (1) are at an early stage of grammaticalisation (These use lexical marking), (2) have developed inflectional future constructions (50 percent of languages belong here), or (3) have low degrees of grammaticalisation (These use periphrastic constructions and have no future tense). English is said to belong to the third category that uses periphrastic constructions.
What is remarkable about these revelations is the equation of tense with inflection. Equally interesting is the suggestion that the grammaticalisation of future tense for English is ongoing and that English will continue to express future time reference periphrastically until the process is completed. In effect, there is no future tense because there is no inflection, and WILL does not mark future tense because it is yet to be transformed into a morpheme. This further supports my claim that grammaticalisation is narrowly interpreted in the literature, and shows that the controversy surrounding the description of tense is not about to end soon, given the recency of the studies reviewed. The point needs to be reiterated that grammaticalisation does not solely entail morphematisation; it does not necessarily culminate in a verb inflection either. While it is acknowledged that language change occurs and that it is natural and gradual, the current state of WILL remains that of a non-morphemic future tense marker. This statement is also valid for its syntactic allies, SHALL and BE GOING TO.
Now that it has been established that future tense in English is marked syntactically by placing WILL/ SHALL or BE GOING TO before the x+o base form verb, it remains to be resolved the true form of BE GOING TO and the issue of whether or not the to-x+o non-finite form after the catenative verb marks future tense. These are taken up next, one after the other.

Is It GOING TO or BE GOING TO?
Described as semantically "a better candidate for the marker of future tense" (Palmer, 1990, p. 161;Palmer, 2001, p. 104) and regarded as "the -ing form of WILL" (Halliday, 1976, p. 149), BE GOING TO is descriptively contentious. There is semantically little difference in meaning between BE GOING TO and WILL/SHALL, except that it refers to "the immediacy of the future action" and, when the reference is to past time, "it means that the event referred to did not take place" (Coates, 1983, pp. 199−200). BE GOING TO is a semi auxiliary verb with idiomatic and quasi modal characteristics. Semi auxiliary verbs are two-or multi-word forms introduced by the primary verb BE or HAVE, followed by a non-finite verb or adjective, and then preposition to. So, they constitute a syntactic unit and function as a syntactic whole because they are a construction (Quirk et al., 1985). BE GOING TO cannot, therefore, rightly be split into two halves for the purpose of tense marking or any other function. The initial BE does not mark tense independently of the entire construction; neither does the lexical verb with preposition to occur or function without BE. However, the systemic description of tense presents BE GOING TO as GOING TO without its initial primary verb element BE in place. This is descriptively anomalous.
There are no known explanations for declaring GOING TO as a future tense marker for English instead of its full syntactic complement BE GOING TO. However, it is not difficult to point at the systemic grammarians' obsession with the notion of recursive tense and their quick desire to demonstrate that this substitute for WILL inherently results in double tense marking, as the paradigm of thirty-six finite verb tense forms show (Halliday, 1976, p. 154;Halliday & Matthiessen, 2004, pp. 340−342). The question that arises at this juncture is the roles that the two parts of the construction BE GOING TO play. From available analyses, the five finite forms of the primary verb (BE) are present (am, is and are) and past (was and were) tense forms, and are therefore markers of primary tenses. In contrast, GOING TO marks future tense at the secondary degree of delicacy. In essence, any tense sequence involving BE GOING TO cannot be analysed as a single tense form because it necessarily entails recursion (The same applies to BE ABOUT TO and BE TO), unlike that realised by WILL/SHALL. This does not subtract from the grammatical fact that BE GOING TO marks future tense as a single syntactic unit relative to the time expressed by its initial BE. The need to account for the specific roles of its component parts should not warrant the commission of the descriptive error which the recognition of GOING TO represents. The following sentential examples are taken from Halliday and Matthiessen (2004, p. 344), with the future tense forms in italics.

The to-x+o non-finite Form as a Marker of Futurity
Those for whom tense distinction is solely morphological (e.g., Greenbaum, 1996, p. 254) state that the first and only finite verb in a verbal group marks tense. Since finite verb forms do not co-occur, it follows that the second, third or nth verb after the finite verb in a sequence of verbal group is non-finite and its status tense-wise is secondary tense. An example is "has been taking", analysed as present-in-past-in-present. But this is not straightforwardly so in respect of future tense and the to-x+o non-finite form. Contrary to widely held views, I hold that the non-finite verbal group also makes tense selections, and that its morphological incompatibility with the known present tense and past tense forms is immaterial. I shall do so by critically examining Comrie's (1985) handling of the tense sequence promised to give.
Coming from a background that conceptualises tense morphologically, Comrie declares that "in English, the infinitive after the verb promise shows no tense opposition" and cites the sentence John promised to give me ten pounds. He explains that the time reference of the to-x+o non-finite form to give can be deduced "to a time subsequent to the time of John's promise", and that this deduction comes from our world view of promises rather than the grammatical system of English (Comrie, 1985, pp. 52−53). This claim is faulty in at least one way. The to-x+o non-finite form expresses "purpose" and "result" as meanings (Strang, 1969, p. 174), and both meanings relate to future time irrespective of the underlying meaning of the catenative verb that precedes it (In the case of promised to give, its meaning is purpose). So, it is not promise alone that makes future time reference in this instance. Indeed, all catenated verbal groups make at least two tense selections depending on whether or not there is an auxiliary verb modifier. What would the interpretation be if the catenative verb were refused, declined or offered instead of promised? (Consider: refused to give, declined to give, and offered to give.) These are verbs without inherent future time reference but which permit the to-infinitive form to follow them in a chain-like manner. It is important to state that, tense-wise, the to-x+o non-finite form is future and that promised to give, like other similar structures, is marked for future-in-past tense. In fact, the catenated verbal group of the Cate + to-x+o structure, which promised to give represents, is the most syntactically productive in English (Adejare, 2013;Adejare, 2016). So, there are two tense selections in promised to give, which are past tense (primary tense) and future tense (secondary tense), realised respectively by promised and to give (Halliday & Matthiessen, 2004).

Conclusion
In this paper, I made the assertion that the controversy surrounding the description of tense in English is traceable to the way tense is conceptualised in the grammar, and linked this to the one-sided interpretation of grammaticalisation, a term that recurs in notable definitions of tense. I demonstrated that the restriction of the term grammaticalisation to mean verb inflection only is responsible for the denial of a future tense system for English and its non-morphological markers, and showed how this results in a partial account of tense that does not correctly reflect everyday language use situation. I subsequently advocated syntactical marking for future tense, showing why it is necessary to recognise both morphological marking and syntactical marking of tense in English. What remains to be done is to highlight the implications of a description of tense that recognises two modes of tense marking-namely morphological marking for past and present tenses and syntactical marking for future tense-for both theory and description. These are outlined as concluding remarks.
a) What has been presented above questions the notion of tenselessness used to refer to languages for which reference to time per se is not "grammaticalised". This is theoretically unsound and descriptively inaccurate and should be revised. The notion of time itself is so culturally important that it becomes inconceivable that some languages would have no means of locating in time. It must be the case then that the so-called tenseless languages are what Dapraetere and Salkie (2015, p. 355) label "languages without inflections of any kind", and that they most probably express temporal location through non-morphological means, such as the syntactical marking established for future tense in English in this study.
b) The syntactical marking of future tense established in this study should cause a redefinition and re-description of tense to take place in order to reflect natural language realities. This should start at the highest level of meta-theory, and move down the line with appropriate degrees of sophistication to descriptive grammar level. Prescriptive school grammars also stand to benefit.